The Effect of Salesperson Behavior and the Influence

Transcrição

The Effect of Salesperson Behavior and the Influence of
Attributional Thinking on Customer Reactions to an
"Open Architecture" Product Offering
DISSERTATION
of the University of St. Gallen,
School of Management,
Economics, Law, Social Sciences
and International Affairs
to obtain the title of
Doctor of Philosophy in Management
submitted by
Winfried Daun
from
Germany
Approved on the application of
Prof. Dr. Torsten Tomczak
and
Prof. Dr. Andreas Herrmann
Dissertation no. 3950
Hundt Druck GmbH, Köln 2011
The University of St. Gallen, School of Management, Economics, Law, Social
Sciences and International Affairs hereby consents to the printing of the present
dissertation, without hereby expressing any opinion on the views herein
expressed.
St. Gallen, October 26, 2011
The President:
Prof. Dr. Thomas Bieger
Meinen Eltern
VORWORT
__________________________________________________
Die vorliegende Arbeit ist während meiner Zeit als externer Doktorand an der
Forschungsstelle für Customer Insight, Universität Sankt Gallen, entstanden.
Parallel hierzu habe ich meine Tätigkeit in Funktionen des Corporate Brand
Management und Marketing bei der UBS AG, Zürich, fortgeführt. An dieser
Stelle möchte ich daher all jenen danken, die zum Gelingen dieser Arbeit
beigetragen und mich in den letzten Jahren unterstützt haben.
Zunächst gilt mein Dank meinem Doktorvater, Prof. Dr. Torsten Tomczak, der
meinem Promotionsvorhaben von Beginn an mit grossem Wohlwollen und
Interesse begegnet ist. Seine konstruktiven Hinweise und klaren Zielvorgaben
waren stets hilfreich und haben ein stringentes Forschungsvorgehen erlaubt.
Ebenso bedanke ich mich bei Herrn Prof. Dr. Andreas Herrmann, der umgehend
bereit war, diese Arbeit als Ko-Referent zu betreuen. Auch mein berufliches
Umfeld hat sehr zum Gelingen der Arbeit beigetragen. Insbesondere gilt mein
Dank dabei meinen Vorgesetzten Bernhard Eggli, Jestyn Thirkell-White und
Daniel Zweifel, die meine berufsbegleitende Dissertation vorbehaltlos und mit
grossem Freiraum unterstützt haben.
Mein ganz besonderer Dank geht fraglos an meinen Dissertations-Betreuer,
Prof. Dr. Daniel Wentzel. Ohne sein grosses Engagement, seine fachliche und
persönliche Unterstützung und seine zahlreichen konkreten Anregungen hätte
die Arbeit nicht in dieser Form entstehen können. Unser freundschaftliches und
humorvolles Miteinander war und ist eine grosse Bereicherung für mich.
Schliesslich möchte ich meinen Eltern von Herzen danken. Sie haben mich nicht
nur in meinem ganzen Werdegang uneingeschränkt unterstützt und gefördert,
sondern mir auch jenes Selbstvertrauen und jene Zuversicht mit auf den Weg
gegeben, die es zur Bewältigung grosser Herausforderungen braucht.
Ihnen widme ich diese Arbeit.
Zürich, im November 2011
Winfried Daun
ABSTRACT
The past decade has seen the emergence of a particular sales and distribution
model in which companies that manufacture products and sell them directly
through proprietary distribution channels open these channels to third parties
and often even competitor products. One of the industries to pioneer this model
has been financial services, where the approach is referred to as “open architecture” offering. This term describes the fact that banks, for instance, do not
only sell their own in-house investment products (such as mutual funds) to
clients, but also those of other companies. Providers of an open architecture
argue that their clients benefit from a wider choice of products and more objective, client-oriented advice. But even though such an extended offering may
come at a price (e.g., loss of market share or the dilution of one’s own product
brand image), there is surprisingly little if any research available on how
customers actually perceive and react to an open-architecture offering.
In order to address this issue, this dissertation investigates if and how customer
reactions are affected when a company sells third-party products next to its inhouse ones. Specifically, the present work examines how customer reactions to
an open architecture are influenced by salespeople behavior and attributional
thinking. To establish a sound understanding of the specific factors and
processes at play, a conceptual model is developed and tested that draws on
research into behavioral cues and attribution theories. A qualitative prestudy
and two experiments confirm the model’s proposition that a salesperson’s persuasiveness of reasoning, the proactiveness in offering third-party products and
the “mix” of in-house and external products have a substantial influence on
customer reactions. Moreover, all three cues are substantially mediated by
customer-oriented attributions. The present research has important implications
for the services and sales literature and it expands our understanding of the
interaction among behavioral cues and customer attributions. Moreover, the dissertation contributes a number of managerially relevant propositions on how to
ensure that an open-architecture offering is successfully delivered to customers.
ZUSAMMENFASSUNG
Im vergangenen Jahrzehnt liess sich die Entwicklung eines spezifischen
Vertriebsmodells beobachten, in welchem Unternehmen, die ihre eigenen
Produkte über proprietäre Distributionskanäle vertreiben, diese Kanäle für
Dritte, oft sogar für Konkurrenzprodukte öffnen. Eine der Branchen, die ein
solches Modell zuerst eingeführt haben, ist die Finanzindustrie, in welcher der
Ansatz als "offene (Produkt-) Architektur" bezeichnet wird. Dieser Begriff
beschreibt den Umstand, dass etwa Banken nicht nur ihre hauseigenen
Anlageprodukte (wie z.B. Investmentfonds) an Kunden verkaufen, sondern
auch solche von anderen Firmen. Anbieter einer offenen Architektur
unterstreichen, dass ihre Kunden von einer breiteren Auswahl an Produkten und
einer objektiveren, kundenorientierten Beratung profitieren. Gleichzeitig birgt
ein derartig erweitertes Angebot aber auch erhebliche Risiken, wie etwa den
Verlust
von
Marktanteilen
oder
die
Beschädigung
der
eigenen
Produktmarke(n). Vor diesem Hintergrund erstaunt es, dass nahezu keine
Forschung zu der Frage vorliegt, wie Kunden eine offene Produktarchitektur
wahrnehmen und auf sie reagieren.
Um diese Thematik aufzugreifen, untersucht die vorliegende Dissertation, ob
und in welchem Masse Kundenreaktionen davon beeinflusst werden, dass ein
Unternehmen Fremdprodukte parallel zu den Eigenen verkauft. Insbesondere
wird die Fragestellung behandelt, ob das Verhalten von Verkäufern und
attributionales Denken von Kunden sich darauf auswirken, wie Letztere auf
eine offene Architektur reagieren. Ziel der Arbeit ist es, zu einem fundierten
Verständnis der spezifischen Einflussfaktoren und relevanten Prozesse
beizutragen. Aus diesem Grund wird ein konzeptioneller Modellansatz
entwickelt und getestet, der auf Forschungserkenntnissen in den Bereichen der
Attributionstheorie und der "behavioral cues" (i.e., Verhaltenssignale) aufsetzt.
Eine qualitative Vorstudie und zwei quantitative Experimente bestätigen die
Hypothese,
dass
die
Überzeugungskraft
einer
Verkaufsperson,
ihre
Eigeninitiative im Anbieten von Fremdprodukten und die "Mischung" von
Eigen- und Fremdprodukten einen wesentlichen Einfluss auf Kundenreaktionen
haben. Darüber hinaus wird die Wirkung aller drei dieser Verhaltenssignale
durch Attributionen von Kundenorientierung mediiert. Die vorliegenden
Forschungsergebnisse haben wichtige Implikationen für die Verkaufs- und
Serviceliteratur, und sie erweitern unser Verständnis der Interaktion zwischen
Verhaltenssignalen und Kundenattributionen. Darüber hinaus gibt die
Dissertation eine Reihe von Management-Empfehlungen, die Unternehmen
dabei helfen können, ihren Kunden eine offene Architektur erfolgreich
anzubieten.
TABLE OF CONTENTS
__________________________________________________
1
2
INTRODUCTION .......................................................................................... 1
1.1
Problem Orientation ................................................................................. 1
1.2
The Case of "Open Architecture" in Financial Services.......................... 2
1.3
Research Questions and Structure of the Dissertation ............................. 7
CONCEPTUAL DEVELOPMENT ............................................................. 11
2.1
The Appeal of Variety............................................................................ 11
2.2
Customer Attributions ............................................................................ 13
2.2.1
Theoretical Foundations .................................................................. 13
2.2.2
Antecedents of Attributional Thinking............................................ 19
2.2.2.1 Disconfirmed Expectations ......................................................... 19
2.2.2.2 Other Triggers of Attributional Thinking ................................... 21
2.2.3
2.3
Outcomes of Attributional Thinking ............................................... 22
Customer Reactions to Salesperson Behavior ....................................... 24
2.3.1.1 Influence of Salesperson Cues on Customer Reactions.............. 24
2.3.1.2 Influence of Cues on Customer Attributions............................... 28
2.3.1.3 Risks of "Priming" Attributions .................................................. 29
2.4
Qualitative Study.................................................................................... 31
2.4.1
Objectives ........................................................................................ 31
2.4.2
Design, Participants and Procedure ................................................. 32
2.4.3
Results.............................................................................................. 32
2.4.3.1 Disconfirmation of Expectations................................................. 32
2.4.3.2 Attributional Thinking................................................................. 33
2.4.3.3 Salesperson Behavior.................................................................. 39
2.4.3.4 Summary...................................................................................... 43
3
HYPOTHESES DEVELOPMENT .............................................................. 45
3.1.1
Impact of Cues on Customer Reactions .......................................... 45
3.1.1.1 Persuasiveness of Reasoning ...................................................... 45
3.1.1.2 Proactiveness .............................................................................. 47
I
3.1.1.3 Product Mix................................................................................. 49
3.1.2
Causal Attributions as Mediator ...................................................... 51
3.1.2.1 Mediation of Persuasiveness of Reasoning ................................ 52
3.1.2.2 Mediation of Proactiveness......................................................... 52
3.1.2.3 Mediation of Product Mix ........................................................... 53
4
EXPERIMENTAL ANALYSES.................................................................. 54
4.1
Overview of Analyses ............................................................................ 54
4.2
Experiment 1 .......................................................................................... 56
4.2.1
4.2.2
Manipulation of Independent Variables .......................................... 58
4.2.2.1 Manipulation of Proactiveness ................................................... 59
4.2.2.2 Manipulation of Product Mix...................................................... 60
4.2.2.3 Manipulation of Persuasiveness ................................................. 61
4.2.3
Selection of Measures...................................................................... 61
4.2.3.1 Dependent Measures................................................................... 61
4.2.3.2 Manipulation Checks .................................................................. 63
4.2.4
Results.............................................................................................. 64
4.2.4.1 Manipulation Checks and Item Reliability. ................................ 64
4.2.4.2 Hypothesis Testing ...................................................................... 65
4.2.5
4.3
Discussion ........................................................................................ 80
Experiment 2 .......................................................................................... 81
4.3.1
4.3.2
Manipulation of Independent Variables .......................................... 82
4.3.3
Selection of Measures...................................................................... 85
4.3.3.1 Dependent Measures................................................................... 85
4.3.3.2 Manipulation Checks .................................................................. 85
4.3.4
Results.............................................................................................. 87
4.3.4.1 Manipulation Checks and Item Reliability. ................................ 87
4.3.4.2 Hypotheses Testing ..................................................................... 88
4.3.5
4.4
Discussion ...................................................................................... 103
Influence of Customer Expertise.......................................................... 104
II
5
DISCUSSION ............................................................................................. 107
5.1
Summary of Results ............................................................................. 107
5.2
Theoretical contribution ....................................................................... 110
5.2.1
Contribution to Literature on Open Product Architectures ........... 110
5.2.2
Contribution to Literature on Cue Influence in Selling................. 111
5.2.3
Contribution to Literature on Customers' Attributional Thinking. 113
5.3
Managerial Implications....................................................................... 114
5.3.1
Delivering an Open-Architecture Offering to Clients ................... 115
5.3.1.1 Sales Force Competence........................................................... 116
5.3.1.2 Sales Force Incentivization....................................................... 117
5.3.2
Promoting an Open-Architecture Prior to the Sales Encounter..... 118
5.3.2.1 Advertising Communication...................................................... 118
5.3.2.2 Brand Positioning ..................................................................... 121
5.3.2.3 Impact of Third-Party Brands................................................... 122
5.4
Limitations............................................................................................ 123
5.5
Future Research.................................................................................... 126
5.5.1
High versus low Customer Expertise ............................................ 126
5.5.2
Manufacturer Role versus Retailer Role ....................................... 128
5.5.3
Perceived Product and Range Fit................................................... 130
6
REFERENCES ........................................................................................... 132
7
APPENDICES ............................................................................................ 151
7.1
Stimulus Materials Used in Experiment 1 ........................................... 151
7.2
Scripts of the Video Treatments in Experiment 2................................ 158
III
LIST OF FIGURES
__________________________________________________
Figure 1-1: Open-architecture print advertisement (2001).................................. 4
Figure 1-2: Print Advertisement (2011) encouraging Banking Clients to ask for
Third-Party Funds ............................................................................ 5
Figure 1-3: Structure of the dissertation .......................................................... 10
Figure 2-1: Examples of attributional antecedents and outcomes ..................... 23
Figure 2-2: Relationship of cues, customers’ attributions and reactions .......... 31
Figure 2-3: Conceptual model of the relationship between salesperson cues,
customers’ attributions and reactions ............................................ 44
Figure 3-1: Effects of salesperson behavior on customer reactions .................. 50
Figure 3-2: Mediation through customer-oriented attributions......................... 53
Figure 4-1: Overview of experimental analyses ................................................. 56
Figure 4-2: Structure of online experiment 1 ..................................................... 57
Figure 4-3: Interaction of persuasiveness and proactiveness in experiment 1;
customer reactions as dependent variables.................................... 68
Figure 4-3 (cont.): Interaction of persuasiveness and proactiveness in
experiment 1; customer reactions as dependent variables ............ 69
Figure 4-4: Interaction of persuasiveness and product mix in experiment 1;
Figure 4-4 (cont.): Interaction of persuasiveness and product mix in experiment
1; customer reactions as dependent variables ............................... 72
Figure 4-5: Interaction of persuasiveness and proactiveness / product mix;
customer-oriented attributions as dependent variable................... 75
Figure 4-6: Mediation of persuasiveness through customer-oriented attributions
in experiment 1 ............................................................................... 76
Figure 4-7: Mediation of proactiveness through customer-oriented attribution
( under conditions of high persuasiveness) in experiment 1 .......... 78
Figure 4-8: Mediation of product mix through customer-oriented attributions
(under conditions of high persuasiveness) in experiment 1 ........... 80
IV
Figure 4-9: Screenshot of the experimental video clip ....................................... 83
Figure 4-10: Structure of online experiment 2 ................................................... 85
Figure 4-11(cont.): Interaction of persuasiveness and proactiveness in
Figure 4-12 (cont.): Interaction of persuasiveness and product mix in
Figure 4-13: Interaction of persuasiveness and proactiveness / product mix in
experiment 2; customer-oriented attributions as dependent variable
........................................................................................................ 98
Figure 4-14: Mediation of persuasiveness through customer-oriented
attributions in experiment 2............................................................ 99
Figure 4-15: Mediation of proactiveness through customer-oriented attributions
(under conditions of high persuasiveness) in experiment 2 ......... 101
(under conditions of high persuasiveness) in experiment 2 ......... 103
V
LIST OF TABLES
__________________________________________________
Table 2-1: Typology of customer-oriented attributions....................................... 35
Table 2-2: Typology of suspicion-oriented attributions ...................................... 38
Table 2-3: Typology of salesperson cues............................................................. 41
Table 2-3: Typology of salesperson cues (cont.) ................................................. 42
Table 4-1: Attribution measurement items.......................................................... 62
Table 4-2: Overview of measures used in experiment 1 ..................................... 63
Table 4-3: Results of multivariate analyses in experiment 1; customer reactions
as dependent variable, main effects .................................................. 66
Table 4-4: Mean values for customer reactions as dependent variables in
experiment 1, main effects................................................................. 66
as dependent variables, interaction effects ....................................... 67
experiment 1, interaction persuasiveness x proactiveness .............. 68
experiment 1, interaction persuasiveness x product mix ................. 71
Table 4-8: Results of a univariate analysis in experiment 1; customer-oriented
attributions as dependent variables .................................................. 73
Table 4-9: Mean values for customer-oriented attributions as dependent
variable in experiment 1.................................................................... 74
Table 4-10: Overview of measures used in experiment 2 ................................... 86
as dependent variable, main effects .................................................. 89
experiment 2, main effects................................................................. 89
as dependent variables, interaction effects ....................................... 90
VI
experiment 2, interaction persuasiveness x proactiveness ............... 91
experiment 2, interaction persuasiveness x product mix .................. 94
Table 4-16: Results of a univariate analysis in experiment 2; customer-oriented
attributions as dependent variable.................................................... 96
variables in experiment 2 .................................................................. 97
Table 4-18: Influence of customer expertise on customers’ attributional thinking
and reactions ................................................................................... 105
Table 4-19: Influence of customer expertise on customers’ attributional thinking
and reactions (mean values) ........................................................... 106
VII
1
INTRODUCTION
1.1
Problem Orientation
Imagine yourself in the following situation. With the intention to afford yourself
a new pair of running shoes, you enter the “Nike” flagship store in your
hometown. After a bit of looking around the heavily Nike-branded interiors,
you’re addressed by Mark, a friendly and sporty Nike shop assistant. He asks
whether he can help you in any way. With a telling glance, you point at the first
unmistakable signs of the potbelly you’ve been cultivating over the last years.
You say “Well you know, I used to run quite a lot, but that feels as if it was back
in the Middle Ages.” You go on telling him about your firm resolution to revive
your old running aspirations. Mark smiles, nods approvingly and you have a
conversation about how often and where you’re usually running. He then leads
you to a gym-like section of the Nike store and lets you do a quick run on the
treadmill. He makes notes on how you set your feet and where you put on the
most pressure. After that, Mark leaves you for a couple of minutes to fetch a
number of different shoes that might suit your needs. When he returns, he puts
four different pairs of running shoes in front of you: Two from Nike, one from
Asics and one from Adidas. He smiles knowingly and says: “The Nike ones here
are great allrounders which I'm sure you'll find very comfortable. But you said
that you're mostly running on tarmac. That’s why I’ve brought the Asics. Their
shock absorption is simply unmatched; you might want to try these out. If they
feel a little too heavy, take a look at the Adidas. They're light-weight.” The guy
seems to know what he’s talking about. But why, you start wondering, would a
Nike shop assistant try to sell you a competitor product?
This scenario seems odd at first, but it is more common than one would think.
The past two decades have seen the emergence of a particular sales and
distribution model in which companies that manufacture products and sell them
directly through proprietary distribution channels open these channels to third
parties and often even competitor products. The form and extent of a
1
collaboration among competitors can vary and its application reaches across
industries as diverse as pharmaceuticals (Dussauge and Garrette 1999), groceries
(Garella and Peitz 2007) and automobiles (Dussauge, Garrette, and Mitchell
2004). And while cooperation in research and development would seem to
represent a typical and (under certain conditions) intuitively sensible case of
competitor alliances (Amaldoss et al. 2000; Hagedoorn, Link, and Vonortas
2000; Hamel, Doz, and Prahalad 1989; Luo, Rindfleisch, and Tse 2007), the
joint distribution of products which are in direct competition with one another
appears slightly unorthodox. One example are traditional own-label retailers that
add well-known manufacturer brands to their assortment (Barr 2009; Garella
and Peitz 2007; Sandler 2009): the British retailer “Marks & Spencer”, a UK
apparel and food retailer renowned for the upmarket quality and positioning of
its products (Sandler 2009) had for decades only been selling food products
under its own Marks & Spencer brand label. On November 5, 2009, that
changed – when the company announced that, for the first time in 50 years, it
would extend its product offering by about 200 “external” product brands, such
as Kellogg’s Cornflakes or Coca Cola (Finch 2009). For British consumers, this
represented a “radical change” (Finch 2009, p. 16) in their grocery shopping
landscape – consequently, the announcement was broadly featured in the news
(e.g., Barr 2009; Felsted 2009; Finch 2009; Sandler 2009). But Marks &
Spencer are far from being the first ones to open their proprietary distribution
network to external products. Few industries in fact have seen a more
widespread adoption of such a model than financial services, where the
approach is referred to as “open architecture” offering.
1.2
The Case of "Open Architecture" in Financial Services
In financial services, the term “open architecture” describes the fact that banks,
for instance, do not only sell their own “in-house” investment products (such as
mutual funds) to clients, but also those of other companies (Kelleher 2007;
Skinner 2006). Over the last few years, the up- and downsides of open product
2
architectures in the financial services industry have been the subject of
considerable controversy. Advocates of this sales model promote several
advantages it is supposed to have. Their “best-of-breed” argument claims that an
extended choice of options improves customers’ chances to get the absolutely
best product for their needs (Narat 2002; Schulz 2002). In line with this point,
banks are said to acknowledge that, however extensive their own product range
is, they cannot always offer the best product in every category and lack the
required specialized expertise (Kelleher 2007). Articles published in the
financial press claim that many financial customers have come to a similar
conclusion and therefore expect their banks to offer also third-party products
(Baum 2005; Severin 2002; Skinner 2006). This seems plausible, given media
headlines such as “Bank-run funds are poor performers”, as proclaimed by the
Financial Times (Johnson 2011). A second client benefit of an open architecture
lies in the promise of greater objectiveness. Pfanner (2002) summarizes this
point by saying that “advisers offer funds and other products from their own
firms as well as competitors, rather than simply pushing in-house offerings." It
is argued that an open-architecture offering de-couples banks’ advisory services
from their product ‘factory’ and thus lends more credibility and perceived
objectiveness to the investment advice that they offer to clients (Kelleher 2007)
The idea is, in other words, that by offering third-party products, a bank’s client
advisor will be perceived as more of a neutral “consultant” rather than a
salesperson. Some banks have quite explicitly played on this argument in their
advertising: German “Commerzbank”, e.g., ran a poster campaign in their
branches that featured the claim “no paternalism, please – third-party funds at
Commerzbank” (Weber 2002). Another German retail bank, Hypovereinsbank,
asked in an advertisement “what else is advice about, if it's not independent”
(HypoVereinsbank 2001, see Fig. 1-1). It does also not surprise that arguments
in favor of an open architecture are strongly supported by individual fund
management companies that are interested in winning banks as distribution
channel for their products – the press contributions of Baum (2005) or
Shaugnessy (2009) would seem typical examples.
3
Figure 1-1: Open-architecture print advertisement (2001)
"My bank offers me independent advice. If another bank has a better investment fund, they
will tell me. They even sell it to me. Makes you wonder what else advice is about, if it's not
independent."
"If another bank has a better investment fund, then my bank will recommend it to me. That
means I don't have to waste my time on running to each and every bank in order to find the
right investment products. That's convenient. I know much better ways to waste my time."
Source: Hypovereinsbank (2011)
4
Figure 1-2 shows a recent print advertising in which Goldman Sachs encourages
banking clients to explicitly ask for Goldman Sachs products (GoldmanSachs
2011).
Figure 1-2: Print Advertisement (2011) encouraging Banking Clients to ask
for Third-Party Funds
“Talk to your advisor about mutual funds from Goldman Sachs” (Goldman Sachs 2011)
The opposite side of the open-architecture controversy is represented by critics
who doubt that the promised advantages of this sales model are genuinely
delivered to customers. They argue that the opening of proprietary distribution
channels has not resolved a major conflict of interest that is posed by the fact
that banks’ in-house products often have the higher margin. “As well intentioned
as open architecture is, it is in-house products that produce revenues", claims
Euromoney magazine (Anonymous 2010). Banks are accused of incentivizing
the sales of their own products (Rasch 2003) and allocating their clients' money
to in-house funds because they do not have to share their margin with a third
5
party (Ross 2010). Bank representatives admit that such a temptation exists
(Speck 2010). If this accusation was justified, the whole value proposition of an
open architecture would seem at stake. As Shaugnessy (2009, p. 6) puts it:
"Worse than having too much or even too little choice is for consumers to be
presented with the illusion of choice where none really exists."
Opponents of an open architecture also point out that an unlimited offering of
similar products can easily overstrain a financial advisor's capabilities and,
consequently, be counterproductive to the goal of recommending the right
solution (Schulz 2002). The latter problem may be resolved by limiting the
amount of third-party providers to a manageable number, an approach often
referred to as “guided architecture” (Baum 2005; Schulz 2002; Wiecking 2003).
However, this approach could undermine the original promise of higher
objectiveness, as it raises the question of how those ‘preferred suppliers’ are
selected and on what basis. Common criticism of guided architectures has
pointed out that fund companies "buy" their way into the distribution channel of
banks through retrocession agreements, contributions to marketing expenses or
revenue sharing (Kelleher 2007; Narat 2002; Schulz 2002; Wiecking 2003).
Consequently, it has been suggested that banks promote those third-party funds
that offer the best ‘kickback’ payments.
A number of points can be concluded from the above excurse. First, open
architecture is a topic of considerable media interest and a subject of ongoing
controversy in the financial services industry. Secondly, as a result of this
discussion, both an open architecture’s potential advantages and banks’
persuasion motives may be salient to at least those clients who have a certain
investment expertise and interest in these matters.
6
1.3
Research Questions and Structure of the Dissertation
The previous sections have argued that there is considerable research available
on collaboration among competitors and that open product architectures in
industries such as financial services are the subject of controversial discussions.
With this in mind, it astounds that information is very scarce on how customers
themselves perceive an open-architecture offering. This is even more surprising
if the obvious risks of such a sales model are considered: If a company opens its
proprietary distribution channel to competitor products, the negative
consequences can be numerous – such as loss of product market share, dilution
of own products’ value proposition and erosion of product brand image. Luo et
al. (2007) warn that too close alliances with a competitor may put a firm's
profitability at risk and lead to the exploitation of its proprietary technologies
and marketing capabilities. With regard to open product architectures in
financial services, banks’ own fund management companies make a good
example of such a negative effect. It seems likely that these in-house product
'factories' will suffer when their exclusive distribution channel is opened to
rivals (Gimbel and Major 2002). Specifically, the fund management units of
banks are anxious that clients will perceive a third-party offering as an
indication that their own products are of inferior quality or variety (Weber
2002). Such potential downsides would need to be outweighed by a strong and
positive overall customer reaction, reflected in parameters such as improved
client satisfaction or purchasing intention. Consequently, retailers highlight their
claim that extending their assortment with top-selling brands offers more
convenience to shoppers (Barr 2009; Sandler 2009), and banks promise access
to 'best in class' products and more objective investment advice (Kelleher 2007;
Narat 2002; Schulz 2002). But is a favorable customer perception of an open
architecture easily achieved, let alone a given? After all, many customers are
wary of sales persons’ ulterior motives and suspect that the advice they receive
is often biased towards the company’s or its salespeople’s own benefit (Bolton,
7
Freixas, and Shapiro 2007; Friestad and Wright 1994; Jonas and Frey 2003;
Krausz and Paroush 2002; Stafford, Leigh, and Martin 1995).
The purpose of this dissertation is to investigate if and how customer reactions
are affected when a company that has up to some point in time only sold its inhouse products starts to sell third-party, even competitor products. It would
appear that the present research is the first to empirically investigate such effects
of an open-architecture sales model. Specifically, the dissertation focuses on the
influence that salespeople behavior has on customer reactions to an open
architecture. To establish a sound understanding of the relevant effects, a
conceptual model is developed and tested that draws on marketing and services
research into behavioral cues and attributional thinking. In order to enhance the
theoretical and practical scope, this model will also incorporate both moderating
and mediating effects, specifying interactions among the independent variables
and describing the circumstances under which the resulting customer reactions
are stronger or weaker. In short, this dissertation aims to answer the following
research questions:
1.
How do customers react towards an open-architecture offering? Are
there specific behavioral cues that salespeople provide during the sales
episode that will influence these customer reactions?
2.
To what extent, if any, is such an influence subject to interaction
effects among the different cues?
3.
Lastly, is the relationship between salesperson cues and customer
reactions mediated by attributions that customers generate in order to
explain the salesperson's behavior?
This dissertation is structured into the following chapters: Chapter 1 presented
an introduction into the topic in general and provided a detailed insight into the
specific context of open product architectures in financial services, as this
industry will provide the background for empirical research. Chapter 2 develops
8
a conceptual model based on existing research into expectancy disconfirmation,
attributional thinking and the effects of behavioral cues. In addition, this chapter
covers the design, procedure and results of a qualitative pre-study undertaken to
substantiate the model's basic assumptions. Chapter 3 then brings forward a
number of research hypotheses based on the conceptual model. The validity of
these hypotheses has been empirically examined in form of two experimental
studies that share the same independent and dependent variables but which
employ different experimental stimuli and sample populations in order to
enhance their practical and theoretical scope. The design, procedure and results
of these quantitative studies are thoroughly documented in Chapter 4. Finally,
Chapter 5 presents a summary of the findings, addresses their implications and
limitations, and identifies potential areas of future research. The structure of the
dissertation is outlined in Figure 1-3.
9
Figure 1-3: Structure of the dissertation
CHAPTER 1: INTRODUCTION
Problem orientation of the dissertation, context information on open product
architecture in financial services and formulation of research questions
pp. 1 - 10
CHAPTER 2: CONCEPTUAL DEVELOPMENT
Review of theoretical foundations, development of a conceptual model and
presentation of supportive findings from a qualitative pre-study
pp. 11 - 44
CHAPTER 3: HYPOTHESES DEVELOPMENT
Formulation of research hypotheses to be empirically tested in two
experimental studies
pp. 45 - 53
CHAPTER 4: EXPERIMENTAL ANALYSES
Documentation of the design and the results of two experimental studies
pp. 54 - 106
CHAPTER 5: DISCUSSION
Detailed discussion of the results, implications, and limitations of the
dissertation and identification of future research
pp. 107 - 131
10
2
CONCEPTUAL DEVELOPMENT
This chapter will build the conceptual foundations that ultimately serve as a
basis for the hypotheses formulated in Chapter 3. It comprises four distinct
sections. In the first one, a short review on the appeal of variety aims to explain
why, prior to all other considerations, many customers may welcome the idea of
an open product architecture. In the next two sections, factors that may influence
their actual reactions towards an open architecture are explored by reviewing
extant literature into expectancy disconfirmation, attributional thinking and the
effects of salesperson behavior. Hereafter, key findings are summarized and
reflected in form of a high-level conceptual model. The fourth and final part
presents results from a qualitative pre-study that substantiate the model’s
assumptions and provide relevant details for the design of subsequent
experimental studies.
2.1
The Appeal of Variety
There are a number of reasons why the fact that a company opens its proprietary
distribution channel to competitor products may have a positive influence on
customer reactions. First of all, the introduction of an open product architecture
offers customers a greater variety of options to choose from. Over the last few
decades, a substantial amount of research has discussed the advantages and
disadvantages of assortment size and variety from a customer’s perspective.
Many of these studies suggest that greater variety of products and services
positively affects customer satisfaction (Broniarczyk, Hoyer, and McAlister
1998; Chernev 2003; Hoch, Bradlow, and Wansink 1999; Kahn and Lehmann
1991) and leads to more positive evaluations of the assortment (Oppewal and
Koelemeijer 2005). One of the main reasons brought forward on this argument
is that a wider range of options provides a higher chance of finding a product
perfectly fitting one's own requirements. Consequently, a greater assortment
11
variety would promise a reduction of perceived risk (Simonson 1990), reduce
the need for alternative-seeking and save time and effort. A second line of
research follows the premise that many customers display a variety-seeking
behavior, actively appreciating the multitude of consumption choices that an
extensive assortment can offer (Chernev and McAllister 2005; Huffman and
Kahn 1998). Building on Kahn and Lehmann’s findings (1998), Hoch,
Brandlow and Wansink (1999) observe that the desire for variety may be even
greater when customers are uncertain what product they need and, consequently,
would prefer. They argue that in such a context, a wide variety promises a
greater flexibility of choice and the chance to form an educated opinion about
the scope of possible solutions. This seems plausible for many product
categories that present the customer with a virtually unmanageable amount of
different, sometimes very complex options, such as investment products,
insurance policies or personalized travel packages. Against this background, it is
hardly surprising that many of the above arguments find themselves in the
reasoning of companies advocating an open product architecture: while retailers
allude to the convenience of ‘one-stop-shopping’ (Barr 2009), financial services
institutions highlight customers’ access to ‘best-in-class’ products (Baum 2005;
Narat 2002).
On the other hand, it has been shown that the positive effect of greater variety
does apply far from everywhere, as too much variety often leads to confusion on
the customer's side and inefficient decision processes (Boatwright and Nunes
2001; Chernev 2006; Greenleaf and Lehmann 1995; Iyengar and Lepper 2000).
Consumers quite plainly 'do not see the woods for the trees'. Such effects,
however, are likely to be ameliorated wherever some form of sales or service
agent acts as intermediate between the client and an overwhelmingly large
choice of products. As long as a bank client, for example, trusts his or her
financial advisor to pre-select the "right" products and thereby significantly
reduce the complexity of choice, a greater choice of products is likely to keep its
appeal. Chernev (2006) argues that customers whose decision process focuses
12
on selecting the right overall assortment rather than an individual product
(because that, in the banking case, would be pre-selected by the advisor) will
prefer larger assortments. In addition, Kahn (1995, p. 139) notes: "Variety may
also be an important consideration when a consumer chooses a portfolio of
options at one time. For example, when choosing financial services or
investments, consumers may choose a diverse portfolio." The latter arguments
are partially based on the premise that intermediates, such as financial advisors
or specialized travel agents, truly use the benefits of their firm’s wider
assortment to their clients’ best interest. How salespeople and advisors would
need to behave in order to give their customers just that confidence is one of the
fundamental questions at the core of this dissertation thesis.
It can be concluded that many customers are likely to appreciate the greater
assortment variety that an open-architecture offers to them – at least in those
circumstances where their product or service selection is assisted by an advisor
or salesperson and as long as they are confident that the salesperson’s preselection or recommendation makes use of the additional variety. Overall, the
promise of an open architecture may therefore tend to instigate positive rather
than negative or no customer reactions. However, clients will rarely be able to
form an opinion about a product or service offering without the influence of any
contextual factors. Instead, it can be assumed that existing beliefs and expectations that the individual holds, as well as the specific circumstances of a sales
encounter, will strongly affect a customer’s attitude formation. These factors
will be discussed in more detail in the following two sections.
2.2
Customer Attributions
2.2.1
Theoretical Foundations
Imagine a scenario similar to the one used in the introduction chapter: The
customer of a high street bank wants to invest money in form of mutual funds.
His or her bank has up to now been selling only products from its own
13
investment fund label. But this time, the bank’s client advisor recommends also
funds from other parties, even competitors. What reaction will the customer
show: surprise, delight or suspicion? First of all, it seems highly plausible that
such a behavior would seem counter-intuitive to many clients. Why, after all,
would someone offer a competitor’s product instead of his or her own? What is
the reason driving such a behavior? In case that no immediate or obvious
explanation is at hand, customers may start looking for a cause.
For many decades, behavioral theory has investigated individuals’ attempts to
understand the events and behaviors that they observe – especially those that
have no obvious explanation. One of the most prominent and enduring
theoretical frameworks is provided by attribution theory, whose foundations
have been contributed by Heider (1958), Jones and Davis (1965), Kelley (1967,
1973) and Weiner (1985a, b). In line with many works of attribution research,
this dissertation will use the term 'attribution theory' (cf. Folkes 1988; Wimer
and Kelley 1982), even though some researchers have pointed out that
'attribution theories' would be more accurate, since the concept is a framework
of different related theories (Kelley and Michela 1980; Mizerski, Golden, and
Kernan 1979).
The term attribution refers to the cognitive process that individuals undertake to
arrive at causal explanations of perceived events (Kelley 1973). Following
Trommsdorff (2009), attribution theory attempts to explain how individuals
infer potential reasons and motives behind observed events and behaviors, but it
also argues how different conclusions subsequently affect the attributing
person's behavior. In line with this definition, Niemeyer (1993) states that
attribution theory deals with effects that are consciously observed by someone
and influence that person's perceptions and actions. The fundamental
assumption of this branch of social theories is that humans strive to understand
the world they live in and the events they witness – as such, attribution theory
attempts to explain the circumstances in which people ask 'why' questions
(Kelley 1973).
14
Going back to the banking example, attribution theory postulates that a client
may perceive the offering of competitor products by a bank's advisor as caused
by very different reasons, e.g.,
a) I am a very demanding and knowledgeable client – the advisor
knows I’d never buy only his own bank’s products
b) The advisor wants to provide me with the best solution for my
needs
c) The advisor is a neutral source of information, and his advice is not
impaired by the fact that he is working for a bank
d) The bank has little competence in this specific investment category
and therefore it resorts to products from third parties
e) The bank gains a high profit margin on external products and
therefore "pushes" them
f) The advisor makes a better commission on the expensive thirdparty products
g) A newspaper on the advisor’s desk features a large advertisement
for a competitor’s award-winning investment funds
h) Open architecture is in discussion all over the financial press and
therefore banks consider it a “must”
i) The bank’s own products are of poor quality
This short list of potential attributions helps to illustrate several fundamental
predications of attribution theory. Firstly, when investigating the potential cause
of an event, consumers can search in different places, commonly termed locus of
causality (Weiner 1985a): They can perceive themselves (a) or others (b, c, d, e,
f) as being the cause. They may attribute the event to the circumstances of a
specific situation (g) or to the subject of the event itself (h, i) (Kelley 1973;
Kroeber-Riel 2003; Niemeyer 1993). In addition to locus of causality, different
works of research have identified a number of dimensions that apply to the
concept of attribution, including those that Weiner (1985a) has summarized as
the controllability and stability of a perceived event (also, Folkes 1984).
15
Controllability refers to the degree to which the outcome of an event is
perceived as being within the control of the entity that caused it (Weiner 1985a).
A customer who waits in a line at the checkout may perceive the checker as the
locus of causality, but acknowledge that the control of the event lies within the
hands of the store manager who refuses to open more checkouts. Similarly, the
offering of third-party funds to a banking client may ultimately not be decided
by the financial advisor, but be influenced by the bank or even different external
fund companies. A more sophisticated judgment like this, however, would
require the customer to be aware of the multiple agents involved in the
“background” of an event: In this context, Folkes (1988) points out that, in spite
of a wide range of responsible entities along the supply chain, a customer may
just distinguish between two causal agents – him or herself on the one side and
the salesperson on the other side. In light of these examples, it does not surprise
that Folkes (1984) has demonstrated that "locus of causality" and
"controllability" are quite similar, reporting a correlation coefficient of .94.
Consequently, Tsiros, Mittal and Ross (2004) merge the two dimensions and
simply talk of "responsibility".
Finally, individuals form an opinion about the stability of the cause that they
have attributed to a certain event, i.e., they assume that the causality behind a
certain event is a common, frequent one or, rather, a one-time, coincidental
occurrence (Folkes 1984; Weiner 1985a). In addition to Weiner's (1985a) main
dimensions of locus of causality, controllability and stability, empirical research
has yielded other attribution dimensions, such as those forwarded by Wimer and
Kelley (1982), which included simple vs. complex or good vs. bad. With regard
to this “valence” of attributions, the present dissertation follows DeCarlo’s
broad distinction between “customer-oriented” and “suspicion-oriented”
attributions (DeCarlo 2005, p. 239). Customer-oriented attributions presume the
influence of altruistic motives, as in the above examples b) or c). "Suspicionoriented" attributions, such as e) or f), reflect different self-seeking motives a
customer might suspect behind a salesperson’s message or behavior (DeCarlo
16
2005). As the term would suggest, such attributions are frequently triggered by
suspicion, a “dynamic state in which the individual actively entertains multiple,
plausibly rival hypotheses about the motives or genuineness of a person's
behavior.” (Fein 1996, p. 1165)
The examples a) to i) help to illustrate more than just the different dimensions of
attributional thinking. They also highlight the fact that in search of possible
causes, customers are likely to develop more than one explanation. Individuals
can attribute a certain event to several causes at a time, depending on its
perceived significance – the more crucial an event appears to the observer, the
more he or she will be inclined to presume multiple reasons (Cunningham and
Kelley 1975). Not all of these reasons may have the same impact on the attitudeformation process (Kelley 1973), but may undergo weighting instead. Coming
back to the banking examples, it seems highly plausible that a customer who is
offered competitor products will entertain multiple assumptions about the
motives behind such an offering. He or she may perceive an advisor’s
recommendation of competitor products as caused by a combination of ulterior
motives on the bank’s and the advisor’s side (e + f), or an interplay of
benevolent advisor behavior and the customer’s own skill (a + b). Moreover, the
different pieces of information that a customer relies on in forming an
explanation may not always be conclusive. Burnkrant (1975) has argued that
consumers in the search for the cause of an observed event are likely to evaluate
several cues that are available rather than only one. However, these cues may
deliver an inconsistent message. A financial advisor, for instance, may 1) listen
carefully to the client, 2) ask thoughtful questions, 3) communicate in a friendly
and open manner, but 4) still seem very reluctant to offer any other investment
funds than his own firm's. To which causes will a client attribute the advisor's
motives in this scenario – is the advisor acting in the client's best interest or does
he want to maximise his own profit? As attribution theory dictates, customers
will favour the inference that seems to be supported by a majority of cues
(Burnkrant 1975).
17
The cognitive processes and dilemmas that attribution theory suggests for a
scenario like the one in which a client is faced with an open product architecture
seem quite plausible. The fact that it describes intuitively understandable human
behavior and reactions may be one of the reasons why attribution theory has
proven to be one of the most resilient and often-used concepts in research on
consumer behavior. As Folkes (1988, p. 548) stated in one of the most
comprehensive reviews of marketing-related attribution research: "Attribution
theory is a rich and well developed approach that has a great deal to say about a
wide range of consumer behavior issues." Consequently, the framework affects
many different disciplines of marketing, including pricing, distribution and
communications (Trommsdorff 2009). While many of the predications of
attribution theory are intuitively plausible and have been corroborated by an
extensive body of research, the concept has also been criticized. Fletcher (1984),
for instance, has argued that the framework represented little but common sense
and therefore contributed few insights of scientific relevance. Weiner’s (1985a)
response was that while individual dimensions and relationships forwarded by
attribution theory may be seen as commonly shared knowledge, it is their
connection and integration within one framework that represents the value of the
theory: “What is not shared knowledge, however, is the conceptual analysis –
the linking of various "understood" empirical relations and the use of similar
principles to explain a vast array of phenotypic observations [...] It is this
systemization, that is, the higher order relations between associations realized in
everyday life, that represents much of this attributional contribution” (Weiner
1985a, p. 570). In addition, some researchers have criticized that attribution
theory overestimated the cognitive activities of consumers (Herkner 1981; cited
in Kroeber-Riel 2003): They argue that many consumer perceptions and
predispositions are not the result of logical and differentiating reflection, as
implied by the concept. In response to this accusation, Wong and Weiner (1981)
have pointed out that attributional research does not claim that individuals are
constantly engaged in attributional thinking or that they consciously strive to
find an explanation for everything they observe. Rather, attributions are
18
instigated in case that existing schemata or convictions fail to explain the
observation in a satisfying manner (Wong and Weiner 1981). In order to shed
further light on this matter, the following chapter will review the antecedents of
attributional thinking.
2.2.2
Antecedents of Attributional Thinking
2.2.2.1
Disconfirmed Expectations
At the beginning of the previous chapter, it was argued that a salesperson’s
offering of competitor products will lead a customer to wonder about the reasons
for such a behavior. However, such a reaction is likely to depend on the
expectations that the customer brings into the sales situation. If the customer is
used to being recommended both own-label and external products, he may feel
little urge to question the motives behind the offering. But if he has rarely (if
ever) been offered third-party products or is generally unaware that such
products are available through the bank’s advisor, an open-architecture offering
will not be in line with what the customer has come to expect. On the contrary,
the client´s original conviction may have been that the salesperson was keen on
selling his or her own firm’s products – a common allegation, for instance, in
financial services (Bender 2009; Bolton et al. 2007). Hence, the offer of
competitor products would seem counter-intuitive to many clients in a similar
situation, and instigate them to ‘wonder why’.
Over the last four decades, a considerable body of research has forwarded the
close relationship between the disconfirmation of individuals’ expectations and
their likelihood to engage in attributional thinking (Hastie 1984; Hunt, Domzal,
and Kernan 1982; Lau and Russell 1980; McPeek and Edwards 1975; Smith and
Hunt 1987; Sujan, Bettman, and Sujan 1986; Wong and Weiner 1981). Some of
these works have investigated the role of an individual’s behavior as the cause
of the disconfirmation, another line of research has investigated the specific
19
impact of communication that violates a recipient’s expectations. The first group
would include the contribution of Pyszczynski and Greenberg (1981) who
showed that unexpected behavior leads to more thorough attributional
processing, a result which is also supported by Hastie (1984). Similarly, Lau and
Russel (1980) found that unexpected results of sports events triggered a greater
number of causal attributions than expected ones. Niemeyer (1993) argues that a
salesperson may actively trigger attributions by surprising a customer with a
positive behavior that is incongruent with what the customer had been
expecting. With regard to communication, Hunt, Domzal, Kernan (1982) argue
that if a sales message contains negative information about a product, this
presumably contradicts customers’ expectations. In response, recipients are
more likely to attribute actual product attributes as the message’s cause and
hence perceive it as more credible. Smith and Hunt (1987) arrived at similar
conclusions. McPeek (1975) also confirmed that unexpected messages increase
the perceived honesty and sincerity of sources, but only suggested that
attributions may play a role. Consequently, marketing researchers have
investigated how the relationship of incongruent messages and favorable
attributions could be exploited in consumer goods advertising (cf. Golden 1977;
Golden and Alpert 1987; Hunt et al. 1982; Pechmann 1992).
Whether expectancy disconfirmation triggers “good” or “bad” attributions is
likely to depend on whether the disconfirmation itself is a pleasant or unpleasant
surprise. It seems only natural that, for example, consumers generate positive
attributions if the waiting time at a supermarket checkout is shorter than
expected, and negative ones if they have to wait longer than they had thought
they would (Tom and Lucey 1995). In summary, a considerable amount of
different studies has shown that the disconfirmation of expectations is one of the
strongest instigators of customers’ attributional thinking.
20
2.2.2.2
Other Triggers of Attributional Thinking
Beyond expectancy disconfirmation, attribution research has identified a number
of other antecedents, some of which could bear a relevance in the context of an
open-architecture product offering. Kelley and Michela (1980) have summarized
attributional antecedents under the broad categories of information, beliefs and
motivation. On a more concrete level, Wong and Weiner (1981) for instance
argue that novel and unknown or stressful events may trigger attributional
thinking, as could a sentiment of frustration. Internal factors like the last one –
antecedents that lie within the attributing person – include motivations or
predispositions of the observer, such as suspicion (Fein 1996), the desire for
control of one's environment (Burger and Hemans 1988) or high involvement
(Niemeyer 1993). Put simply, involvement can be defined as the feeling of
personal engagement or interest a consumer has in a stimulus (Greenwald and
Leavitt 1984; Zaichkowsky 1985). For the present context of financial services,
Aldlaigan and Buttle (2001) have demonstrated that financial customers' usage
of investment services is a high involvement activity – unlike, for instance, the
usage of a cheque book – , even though involvement may vary across different
types of customers (Howcroft, Hamilton, and Hewer 2007). Other findings from
attribution research would seem to support the basic notion that financial
investments provide a context for attributional thinking: in case that the
investment is of considerable size and potential losses would substantially affect
the investor's financial situation, making the right investment choice is likely to
be an activity of great personal importance. It is such stressful or simply highimportance events and conditions of "cognitive unrest" (Lichtenstein and
Bearden 1986, p. 295) that have also been claimed to be antecedents of
attributional thinking (Lichtenstein and Bearden 1986; Wong and Weiner 1981).
To conclude, the confrontation with an open-architecture offering may not only
instigate customer attributions because it disconfirms expectancies. Other
factors such as stress or frustration, suspicion, desire for control or high
involvement may also increase a customer’s perceived need to search for causes.
21
2.2.3
Outcomes of Attributional Thinking
As the previous sections outline, individuals use attributions not only to identify
“who or what is responsible” for a certain event (locus of causality,
controllability) and whether it is likely to reoccur (stability). Attributions also
affect the attributing person’s behavior, triggering a response (Niemeyer 1993;
Trommsdorff 2009). Such consequences seem obvious: after all, the valence of
attributions can differ significantly, ranging from positive conclusions about the
benevolence, sincerity or honesty of a communicator or the quality of a product
to negative assumptions about incompetence, ulterior motives or flawed design.
A vast body of research has demonstrated the impact of attributional thinking on
customer behavior, with service failure and recovery being one of the most frequent subjects (cf. Casado Diaz and Más Ruíz 2002; Chebat et al. 1995; Curren
and Folkes 1987; Folkes, Koletsky, and Graham 1987; Hess Jr, Ganesan, and
Klein 2003; Iglesias 2009; Swanson and Kelley 2001). Folkes, Koletsky and
Graham (1987) have demonstrated that consumers’ attributions affect their
propensity to complain and their likelihood of repurchasing a product. Chebat et
al (1995) show a similar relationship between attributions and customer
perceptions of service quality. The importance of “who is responsible” is
underlined by findings of Curren and Folkes (1987) who found evidence that
locus of causality and controllability (i.e., responsibility) affect consumer’s
willingness to communicate – independent of attribution valence. For negative
as much as positive attributions, customers were more likely to talk about a
product when they perceived the seller as responsible for the observed event
(product failure vs. product success). The effect of control attributions on
repurchasing and complaining intention is also supported by Casado and Mas
(2002). Swanson and Kelley (2001) have shown that buyer’s likelihood to
provide word-of-mouth increases when the attributed cause is perceived as
stable. Iglesias (2009) argues that attributions have both a direct and indirect
impact on customer satisfaction. For the case of service failure, the author
showed that customers who perceived the service provider as responsible
22
displayed lower levels of satisfaction and lower evaluations of the service's
overall quality and less perceptions of the service's individual attributes. Similar
results were found by Tom and Lucey (1995) who demonstrated that in case of a
waiting time that is shorter than expected, consumers are more satisfied when
the cause of the shorter waiting is perceived as under the store's control or when
the cause was perceived as stable. The effect was mirrored in case of negative
confirmation, where perceived control and stability led to greater dissatisfaction.
In summary, there is ample evidence that attributional thinking affects customer
reactions such as satisfaction, willingness to communicate, propensity to
complain and intention to (re-)purchase. Moreover, they determine expectations
towards further action – if a firm is seen as responsible for a product failure, it is
expected to apologize (Folkes 1984) or to provide a certain reparation (Hess Jr
et al. 2003).
More than three decades of scientific contributions underline how relevant
learnings from attribution research are for marketing and sales, and especially
for personal selling encounters (Johnson 2006). They bear fundamental
implications
for
salesperson-consumer
encounters
(Niemeyer
1993;
Trommsdorff 2009) as much as for negotiations between professional sellers and
buyers (Folkes 1988). Figure 2-1 summarizes the relationships outlined in
chapter 2.2.
Figure 2-1: Examples of attributional antecedents and outcomes4
Antecedents
• Disconfirmed
expectations
• Suspicion
• High involvement
Attributions
•
•
•
•
Benevolence
Credibility
Competence
Ulterior motives
23
Consequences
• Satisfaction
• Intention to
purchase
• Intention to
recommend
Up to this point, it has been argued that the very existence of an open product
architecture may trigger customer attributions and, subsequently, reactions:
Firstly, because it carries the appeal of variety. Secondly, because it disconfirms
customers’ expectations of how firms advocate their own products, which leads
to favorable attributions. These in turn influence customer behavior. However,
how customers react to a certain product or service offering is also likely to
depend on the agent that “connects” a customer with an offering. The following
chapter will therefore review to what extent salesperson behavior can influence
customer reactions to an open product architecture.
2.3
Customer Reactions to Salesperson Behavior
2.3.1.1
Influence of Salesperson Cues on Customer Reactions
In many sales scenarios in which a customer seeks a salesperson's help in
picking from a large, often intransparent range of options (e.g., in financial investment planning, travel booking or insurance policies) the salesperson can act
as a gatekeeper between a company's overall product offering and the customer:
the salesperson has an overview of the available options that the customer often
doesn't have. In the case of a bank's open architecture, it may be entirely up to
the bank's financial advisors to offer third-party products to their clients – or not.
And even if they did, there would be many ways of positioning the competitor
products in a way that the bank's own products seemed preferable. Many customers are aware of this dilemma and the different tactics that salespeople apply
to persuade them of a certain choice (Campbell and Kirmani 2000; Friestad and
Wright 1994; Niemeyer 1993; Trommsdorff 2009). One of their response
strategies is to look for certain cues in a salesperson's behavior that will reveal
his or her “true intentions”, or at least indicate whether the salesperson strives to
act in the customer’s best interest or is merely focused on selling.
24
Researchers have investigated an extensive range of different salesperson
behaviors in order to measure salespeople's customer- and selling-oriented
behavior, both from a seller's perspective (Saxe and Weitz 1982) and a customer
perspective (Michaels and Day 1985; Wood et al. 2008b). Salesperson behaviors
such as accurately describing the features and benefits of a product and
providing all the information a customer has asked for reflects customerorientation (Michaels and Day 1985). An insurance agent, for instance, who
gave a very clear overview of several life insurance products, highlighting their
different fee structures and payout conditions and explaining their investment
approaches, would be perceived as customer-oriented. Overselling a product or
stretching the truth indicates a strong seller-orientation (Michaels and Day
1985). In the life insurance example, the agent could display such behavior by
describing a mediocre, expensive insurance policy as "great" or downplaying the
financial risks inherent in a life insurance whose capital was mainly invested in
stocks. In a similar vein, Wood, Boles and Babin (2008a) have explored “good”
and “bad” cues that are perceived by customers. Beyond those related to a
salesperson’s behavior, they list other aspects of a firm’s or a salesperson’s
appearance that affect a customer’s opinion of a business – such as the
cleanliness and professional “look” of the business, and the gender, age, race
and outfit of the sales agent.
The cues that customers perceive often have a consequence. Salesperson behavior that reflects customer-orientation or selling-orientation has been shown to
affect customer satisfaction with the salesperson, but also with the product and
its manufacturer (Goff et al. 1997). Such a ‘blanket judgment’ would seem in
line with findings from Crosby, Evans and Cowles (1990), who claim that
consumers often do not distinguish between salesperson and selling firm. Over
the last 30 years, many studies have highlighted how strongly cues affect critical
aspects of a selling encounter. They demonstrate a substantial effect of cues on,
e.g., the formation of trust (Doney and Cannon 1997; Hawes, Mast, and Swan
1989; Kennedy, Ferrell, and LeClair 2001; Sirdeshmukh, Singh, and Sabol
25
2002; Swan, Bowers, and Richardson 1999; Wood et al. 2008a; Wood et al.
2008b), the attribution of credibility (Pornpitakpan 2004), the persuasiveness of
sales messages (Fein 1996; Sparks and Areni 2002) and selling effectiveness in
general (Weitz 1981; Weitz, Sujan, and Sujan 1986). Among these
contributions, a rather common finding has been that the trustworthiness of a
salesperson is, to some extent, inferred from cues that indicate likeability and
competence (cf. Hawes et al. 1989; Swan et al. 1999; Wood et al. 2008a).
According to Buda and Zhang (2000), cues that convey credibility will lead
consumers to take a sales message rather at face value than counter argue it: A
shop assistant in a golf equipment store, for instance, is more likely to persuade
his customer of a specific club’s superiority if he turns out to be a passionate,
experienced golfer himself. Importantly, a considerable body of research
suggests that buyers assess certain characteristics of salespeople and make inferences about the ulterior motives driving their behavior (e.g. Campbell and
Kirmani 2000; Fein 1996; Wood et al. 2008b). In other words, salesperson cues
may represent a critical input for attributional thinking; an implication that will
bear further relevance in the development of the conceptual model.
Many of the above studies have categorized cues in slightly or substantially
different ways. Based on an extensive review of the variables that drive
credibility and message persuasiveness, Pornpitakpan (2004) makes a distinction
between "source variables" such as physical attractiveness or gender and
"message variables" such as argument quality or message discrepancy. In their
analysis of the different communication variables that drive message persuasiveness, Areni and Cox (1995) distinguish between arguments and cues. According
to their definition, an argument is "a communication element that constitutes
part of the fundamental message, and cues […] refer to a communication
element that creates a setting for the message" (Areni and Cox 1995, p. 198).
The two authors admit that the clear separation of these concepts is difficult,
since arguments and cues can take on multiple roles in the persuasion process.
Swan, Bowders and Richardson (1999) argue that determinants of trust can be
26
separated into “direct” and “indirect” cues, the former representing trustinducing behaviors (e.g., revealing one’s own commission fees), the others sales
person features (e.g., likability, outer appearance) that customers use to attribute
trustworthiness. Their findings suggest that direct cues tend to have a stronger
effect, especially over time. While some of the above cited studies may differ in
their categorization of cues, they all support the idea that cues have a substantial
impact on customer reactions.
In addition to the different qualities or categories of cues that a client may perceive, his or her judgment is likely to be formed not only on impressions gained
during a specific sales encounter. It will often be influenced by existing beliefs
or other factors (such as word-of-mouth) accessible to the client in advance of
his actual encounter with the sales agent. The present research, however, focuses
on direct, behavioral, and “in situ” cues, e.g., the ways in which a salesperson
acts when presenting and arguing during a sales episode. This is done for several
reasons. Firstly, it has been argued at the beginning of this section that a
customer’s choice of products often depends on the pre-selection done by the
salesperson – especially, where the complexity of an open product architecture
is given. Being aware of this dependency, it seems highly plausible that a
customer will carefully monitor the salesperson’s behavior, draw inferences on
the agent’s intentions and determine an appropriate reaction. Secondly, it seems
fair to assume that “in situ” cues (though not all) are more easily manageable by
the company and its sales force than some of those observed outside of the sales
encounter – the advice given by a family member, for instance. Hence, the focus
on “in situ” cues promises results of greater managerial relevance. Lastly, a
focus on certain explicit behaviors allows for the investigation of cues that are
relatively specific to the scenario of an open-architecture offering. As such, they
may deliver more novel and theoretically relevant insights than indirect cues
such as outer appearance or likeability of a salesperson.
27
2.3.1.2
Influence of Cues on Customer Attributions
The above chapter forwards the idea that salesperson cues have an influence on
customer reactions. This effect may not (always) be of a direct nature, though.
Rather, salesperson cues will often provide an input on which customers draw
causal inferences about the motives behind a certain behavior (Campbell and
Kirmani 2000; Fein 1996; Wood et al. 2008b) which then affect their reactions.
Given the earlier insights into the function of customer attributions, their causes
and effects, it seems very plausible that attributional thinking forms part of the
cognitive processes that take place between the observation of a cue and the
instigation of a reaction. In other words, salesperson behavior may trigger
attributions or at least influence which attributional causes an individual
perceives as likely.
Consumers have been found to evaluate a multitude of available cues when
searching for the cause of an observed event (Burnkrant 1975). Those will
frequently include aspects of a salesperson’s behavior, but are not limited to it.
Su and Tippins (1998) argue for the case of product failure, that customers'
attributions on "who is to blame" within the supply chain (manufacturer or
retailer) are significantly affected by brand visibility. They conclude from their
findings that well-known brands are less likely to be attributed as the original
cause behind a product failure. In other words, the brand acts as a powerful cue
that refutes the notion that the product failure was the manufacturer’s fault.
Wood, Boles and Babin (2008a) demonstrate that, in order to establish a
salesperson's credibility, a client will draw on a number of cues gained from
both the salesperson and the selling environment. They suggest that attributions
mediate the relationship between cues and customer attitude: "Given that cues
influence a respondent's attributions about trust-building characteristics, study
results indicate that these assessments initiate a process leading to trust of the
salesperson and the firm" (Wood et al. 2008a, p. 34). DeCarlo (2005) has
explicitly shown that such a mediating effect exists. In addition, Rose and
Dickson (1988, p. 106) argue for negotiations between professional sellers and
28
buyers that their research findings “are certainly supportive of the proposition
that attributions mediate the effect of one bargainer's behavior on the behavior of
his bargaining partner”.
The idea that certain cues instigate attributions which in turn affect customers’
behavior has led researchers to suggest that specific cues should be provided
deliberately in order to elicit a desired consumer behavior. Trommsdorff (2009,
p. 291) argues that “it is obvious for personal selling to use attributional aids in
order to shape the perception of a salesperson as competent, altruistic, solutionoriented and credible.” In light of the negative effects that attributional thinking
can potentially have on customers' evaluation of a service encounter, Hui and
Toffoli (2002, p. 1841) conclude that "service managers should try to encourage
customers to form attributions in positive service encounters and to inhibit […]
or manipulate consumer attributions (against the natural attribution biases) in
negative service encounters." Rose and Dickson (1987) have shown how cues
affect impression formation and attribution in a bargaining context. They argue
that “an understanding of the factors capable of stimulating inference-making in
bargaining might improve bargaining skills by suggesting tactics that would tend
to elicit certain types of attributions and impressions favorable to the tactician”
(Rose and Dickson 1987, p. 382).
2.3.1.3
Risks of "Priming" Attributions
Promising as they seem, the above described manipulations may neither be easy
to bring about nor free of risk. Deliberately providing a certain cue in order to
activate a consumer's use of specific (and favorable) decision heuristics is
commonly termed as “priming”. Priming, in the definition of Stafford (1996, p.
37), is based on the idea that sellers can "surreptitiously induce favorable
decision rules in buyers in order to further the sale". The effectiveness of
manipulating consumer through priming social information cues is likely to be
limited, though. Stafford (1996) argues that any sort of ‘subtle’ priming must be
29
unreliable since its effect will strongly depend on the seller's communication and
presentation skills as well as on the buyer's wariness and persuasion knowledge.
Obtrusive priming, on the other hand, carries the substantial risk that consumers
detect the deliberate persuasion attempt behind it and their perception is affected
negatively. This thesis is backed by persuasion research from Friestad and
Wright (1994), who reason that many customers are well aware of persuasion
attempts and will counteract against them. Typical reactions will include active
counterarguing, bolstering one’s own initial attitudes, derogating the message
source, avoiding incongruent information or blaming the experienced anger or
irritation on the message or its source (Tormala and Petty 2004). In addition to
the risks of obtrusive priming, Sharma (1996) has questioned the transferability
of priming concepts to sales situations in general, arguing that situational
parameters in a sales context are critically different from the setup of sociopsychological experiments.
Up to this point, chapter 2 has reviewed several streams of research that
contribute to building a conceptual model for this dissertation. A first conclusion
is that an open-architecture offering is likely to be perceived favorably by
clients. It promises greater choice and variety, and with that, a higher likelihood
of finding the right product. Moreover, a firm that offers competitor products
side by side to its own products seems to act against the obvious selling motives
that many customers are wary of. Their expectations are, consequently,
disconfirmed in a positive way, which should lead to customer attributions in
favor of the firm. Attributions in turn have a substantial influence on customer
reactions such as purchasing intention or willingness to recommend. While these
arguments support the idea that an open architecture per se triggers attributional
thinking, a customer is likely to draw on several sources of information when
forming an opinion about a product offering. The context of a sales episode, and
especially the behavior and appearance of the salesperson, should provide cues
that constitute such information. Not only is there abundant evidence of a strong
relationship between cues and customer reactions, cues have also been found to
30
instigate and affect attributional thinking. Hence, in their search for the reasons
behind an open-architecture offering, customers will evaluate different cues that
are available to them, infer likely causes and react accordingly. In short,
different cues should affect a customer’s reaction to an open product architectture, and this influence is likely to be mediated by attributions. Figure 2-2
summarizes the arguments that have been developed until this point in form of a
high-level conceptual model.
Figure 2-2: Relationship of cues, customers’ attributions and reactions5
Customer
attributions
Cues obtained
from salesperson
behavior
2.4
Qualitative Study
2.4.1
Objectives
Customer
reactions
While the relationships outlined in the previous sections are based on substantial
evidence generated by marketing and services researchers, little is known on
how customers specifically perceive an open-architecture offering and how they
are likely to react to it. A qualitative study was therefore undertaken to further
substantiate and concretise the assumptions of the proposed model and provide
relevant details for the design of subsequent experimental studies.
31
2.4.2
Design, Participants and Procedure
A series of semi-structured in-depth interviews was conducted with Swiss and
German banking clients in the time between March and May 2009. A sample of
10 male and 10 female interviewees was recruited, their ages ranging from 30 to
78, from a wide range of socioeconomic backgrounds (‘standard’ retail banking
clients with financial assets below CHF 25k to “high net worth” individuals with
assets exceeding CHF 1m). The interviewees were told that they participated in
a research study on their general experiences with investment advice and
products offered from their bank. In order to appropriately incentivize the
interviewees for talking about sensitive matters such as personal finance, the
introductory email informed them that 50 CHF would be donated to a charity
project for each interview. In a first step, they were encouraged to talk about the
experiences they had made with investment advice in general, whether and how
they had received advice from a bank’s client advisor and what kind of products
had been discussed. In almost all cases, the respondents also talked very
explicitly about the expectations and general convictions they had brought and
would bring into such an advisory meeting. The discussion then went on to
address whether their bank had ever offered them third-party products, how they
had perceived this or, if not, how they would react to such an offering. In the
final stage, the respondents were asked to indicate conditions under which they
would appreciate or resent being offered third-party products. Each interview
was recorded and lasted from 40 to 70 minutes. After transcription and
categorization of the interviews, a content analysis was conducted.
2.4.3
Results
2.4.3.1
Disconfirmation of Expectations
Results from the interview series demonstrate that many respondents held the
expectation that a bank would only offer them its own investment products to
32
choose from; for many of them, this expectation was rooted in personal
experience. A few exemplary quotes are listed below:
"I would not have expected of an advisor of [bank name] that he would
present other products than his own." (female, 39)
"I would not go to a bank in order to get advice, because, well, I would
always have the feeling they offer me only their own products." (female, 37)
"Well, it's clear, they prefer their own products...I would assume they have
the biggest margin on their own products." (male, 59)
"With funds from banks, I tend to be suspicious, because I always presume
that they will foist their own funds on me." (female, 38)
"I simply had this idea that a bank's advisor is not objective." (male, 59)
2.4.3.2
Attributional Thinking
The interview results suggest that the above described beliefs and convictions
were disconfirmed by the idea an open-architecture offering. They also indicate
that attributional thinking was triggered. The vast majority of respondents
(n=19) articulated one or several motives they would suspect behind a
recommendation of third-party products. Several interviewees explicitly
mentioned that such an offering would make them ‘wonder why’ (n=4):
"I would ask him why.” (male, 53)
"I would ask 'Why?' Why does he offer me third-party products? If he's
offering only his own products, well, that makes sense - he's an employee of
the bank, after all. Questions would come up...it cannot be that he's just a dogooder!" (female, 37)
"Then I would ask him why!" (male, 64)
"...that would raise suspicion in me. I would need to ask: Why? Why does he
offer this to me? What is his stake in this?" (female, 37)
33
The study revealed a number of different attributions, which, as presumed, can
be separated into customer-oriented and suspicion-oriented ones. The most
frequently cited favorable attributions were that the bank was trying to offer a
more complete and sophisticated range of products (n=7 respondents), that the
advisor had the client’s best interest at heart (n=5) or gave neutral and objective
advice (n=4). The majority of respondents would perceive an open-architecture
model as a sign of (more) neutral and objective investment advice which in turn
indicates that the bank has its clients’ best interest at heart. They tend to take the
advantages of a wider choice of products at face value. Some interviewees
underlined this perception with statements like "this shows me that [the advisor]
has an intention that I'm doing well" (male, 59) or "it would have felt like there
is somebody […] committed to get the best out of it for me" (male, 43). A fairly
similar attribution is that of "neutrality" or "independence" as a cause behind the
third-party offering. In the interviews, such a reasoning was suggested by
respondents claiming they would "like to have this kind of neutral advisor, and
this would support the impression" (male, 37) or that they would "probably
perceive it as relatively independent advice" (female, 36). By recommending
external products, the advisor also indicates "that he knows the market and has
an overview of what's going on" (male, 37) and "that he's familiar with other
products" (female, 37). Such statements highlight that some clients will attribute
an open-architecture offering to high competence and a superior market
overview on the advisor's side. One of the most frequently mentioned causes is
the assumption that no bank can cover all possible investment solutions with its
own product portfolio and therefore needs to seek partnerships. It is important to
note that all respondents who arrived at this specific attribution saw it as a
logical consequence of the market’s diversity and not as a weakness of the
bank’s in-house offering. The different customer-oriented attributions are listed
in Table 2-1.
34
Table 2-1: Typology of customer-oriented attributions 1
Attributions Definitions
Client assumes that
Advisor is
independent/ the advisor‘s behavior
is caused by a
objective
„neutral“ position,
i.e., one that is not
biased by any ulterior
motive
Examples
I'd like to have this kind of neutral advisor,
and this would support my impression.
In that case, I would probably perceive it as
relatively independent advice.
It would surely reinforce his independence.
Client attributes the
advisor‘s behavior to
a strong focus on
client needs and a
commitment to place
clients’ interest ahead
of self interest
This shows me [...] that he […] has an
intention that I'm doing well.
Advisor has
strong
product
expertise
Client perceives the
advisor‘s behavior as
a sign of product
expertise
With this, he would demonstrate that he
knows the market and has an overview of
what's going on.
Bank
extends its
offering
bandwidth
Client attributes the
product selection
offered to him to
„gaps“ in the bank‘s
own product range
and to the bank‘s
intention to close
these gaps (by
including third-party
products)
I think, even a bank X might not be able to
cover the whole market of products that exist
in the financial world.
Advisor has
client's best
interest at
heart
It shows me that in this moment, my client
advisor is busy covering my
requirements...and not those of his sales track
record.
To me, it would have felt like there is
somebody […] committed to get the best out
of it for me, not only [selling] his own funds
"by default".
It would also tell me that he's familiar with
other products, not only the ones from his
own bank.
You can't be the best in every area. Therefore,
you need to cooperate with people in places
where you have some weak spots...I don't
think there's anything infamous to it.
I think, in some parts they probably have to
do this, because they are not able to provide
the full bandwidth of products […] on the
specialized side, they probably have to add
something from external.
35
Negative attributions were much less frequent and included suspicions that the
advisor’s product recommendation was driven by obscure inter-company
‘alliances’ (n=4) or ulterior motives such as profit-making (n=1). Overall, few
interviewees established a direct link between the offering of external products
and unfavorable causes. Some respondents who had expressed a strong
suspicion towards banks in general – supported by their own bad experiences –
attributed the open architecture to "hidden alliances" of obscure intention. As
one respondent put it, "a small doubt would remain...who's dealing here with
whom, who's pushing products across to whom?" (female, 59). Another
expressed his anxiousness that "there's some form of alliance or partnership in
the back-ground, of which I do not know" (female, 36). For these clients, the
open architecture did not come across as a sign of neutrality. This is in line with
what Fein (1996, p. 1196) assumes: “Once they have become suspicious,
perceivers are more likely to perceive a wide range of behavior as attributionally
ambiguous. This mind-set may stem in part from the desire not to be duped by
another individual.” Other statements indicate that clients suspect ulterior
motives like the maximization of sales commissions or the achievement of sales
targets: "Maybe there's also an interesting incentive on the financial side, for
external products." (female, 36).
Very few interviewees perceived the offering of third-party products as a sign
that the advisor does not believe in his/her own products or that the bank's inhouse products are of inferior quality: "Does your fund underperform that much
that you're not offering it to me?"(female, 38). However, this attribution was
only triggered by a significant alteration of the question that respondents were
confronted with – namely, when they were asked how they would perceive it if
the advisor offered them only third-party products and none from his or her own
bank. This limitation is critically important. As it has been pointed out at the
beginning, banks’ in-house product factories tend to argue against an openarchitecture model: They claim that clients will perceive such an offering as a
36
confession that the bank’s own products are of inferior quality or variety. The
interviews did not yield any evidence of such an overall, undifferentiating effect.
Finally, the data also supports the idea that clients might perceive an open
architecture as a sophisticated form of persuasion tactic. As one respondent put
it: "[My suspicion would be] that in times of need, banks cooperate. And they
go: 'People appreciate it if you also offer products from other companies.
Hence, if you offer our products, we'll offer yours'" (female, 37). This would
indicate that persuasion knowledge, as Friestad and Wright (1994) have
described it, is being applied. Such a conclusion is also backed by the fact that a
majority of respondents suspected that their bank’s actions were generally
driven by ulterior persuasion motives such as “achieving sales targets” or
“maximizing profits”. At this point, it is important to note that unfavorable
attributions do not always entail an overall negative evaluation (Campbell and
Kirmani 2000). The qualitative data would seem to confirm this impression.
Even those clients who said they were aware of the bank’s sales motives
claimed they would appreciate the market overview and competence displayed
by offering third-party products. Those few clients who were either highly
familiar with the idea of an open architecture and/or had a long-lasting, trusted
and friendly relationship with their advisor were less prone to see any
unfavorable ulterior motives behind their bank’s offering. Again, this is in line
with Friestad and Wright (1994) who claim that persuasion knowledge is not
always and necessarily accessed, but that occasionally, existing knowledge
about a certain topic or about the sales agent himself might dominate the attitude
forming. The different suspicion-oriented attributions are listed in Table 2-2.
37
Table 2-2: Typology of suspicion-oriented attributions2
Attributions Definitions
Advisor has
ulterior
motives
Bank has
ulterior
motives
Client assumes that
the advisor‘s behavior
is caused by ulterior
motives such as
maximising his own
return from
commissions or
achieving a certain
sales target
Client perceives the
product selection
offered to him as a
sign of ulterior
motives on the bank‘s
side, such as the
maximization of fees
Bank's own Client attributes the
products are product selection
offered to him to the
inferior
low quality of the
bank‘s own product
range
Examples
He probably makes more money on these,
right? Even more!
Maybe there's also an interesting incentive on
the financial side, for external products.
Of course, there's also the question of how
they have been incentivized by the other banks
and what the commissions are like.
I believe it's the internal quota that's driving
him. Yes, I think the advisor also gets his
orders from above, like "you have to place so
many millions of these per year."
I don't know...a small doubt would
remain...who's dealing here with whom, who's
pushing products across to whom?
I would think there's some form of alliance or
partnership in the background, of which I
don't know.
I guess my first reaction would be: "Well,
have you struck an alliance? How many sales
orders need to get signed?"
I'd have my own thoughts, whether the bank
itself is very effective then.
[I would ask] does your fund underperfom
that much that you're not offering it to me?
Then I would have some questions about the
bank's own products.
[I'd ask myself]…are their products so
terrible?
Maybe the [third-party] products are better.
38
2.4.3.3
Salesperson Behavior
A second objective behind the qualitative study was to provide a first answer to
the question whether customers’ perceptions of an open-architecture offering are
influenced by certain behavioral cues. It seems likely that clients will pay
specific attention to cues that seem to support their initial inferences – research
has shown that people often selectively absorb information that supports their
initial beliefs and opinions (Jonas et al. 2001). Consequently, they will interpret
available cues as reflecting a salesperson's expertise, ulterior motives,
benevolence or other traits and motives. In the present research, such cues would
mainly include the appearance, behavior and sales messages provided by a
financial advisor. Indeed, the interviews revealed a number of cues provided
during the sales episode that seem to have an influence on whether unfavorable
or favorable attributions prevail in the client's attitude-forming process. This
would include behaviors and statements that very directly suggest honesty,
benevolence or expertise. This would be the case if, for instance, the bank
advisor admitted that he or she earned a commission on selling a specific
product or advised against buying a specific in-house product because its
performance was unsatisfactory. Similar cues have been researched quite
extensively (cf. Crosby et al. 1990; Doney and Cannon 1997; Sirdeshmukh et al.
2002; Swan et al. 1999; White 2005; Wood et al. 2008a; Wood et al. 2008b).
Therefore, this dissertation focuses on cues that are more specific to the scenario
of an open-architecture offering, as they promise both relevant and novel
insights.
Three distinct cues present in an open-architecture context have emerged from
the qualitative data and are listed in Table 2-3. Firstly, clients appreciate an
advisor’s proactiveness in offering third-party products (n=3): statements like
“the advisor should put them (the third-party products) on the table at the same
time as the in-house products” (female, 38) suggest that clients want their
advisor to treat the bank’s own products and external ones equally and to
provide a transparent overview. If customers have to explicitly ask for third39
party alternatives, they are prone to suspect that the advisor intended to withhold
these options from them. As one interviewee put it: “Well, that would seem like
a sales-force approach to me: It would tell me that this guy (the advisor) wants
to see first whether I’ll take what he’s got in-house.” (female, 38)
A second cue that emerged from the interview data is the perceived ‘balance’ of
the product mix, as it is reflected in the share of in-house vs. third-party products
within the advisor’s recommended selection (n=9). Many interviewees
demonstrated a clear expectation towards their bank: They want to see both own
and third-party products getting offered to them. For some clients, it is
acceptable (and somewhat foreseeable) that the share of in-house products may
be more than 50% (“I’d expect 2/3 in-house, 1/3 third-party; male, 53), as long
as external options do not become a mere side note. With such an offering, the
bank proves its competence as both product manufacturer and – relatively –
neutral advisor.
Finally, the most frequently mentioned cue is the persuasiveness of reasoning
provided by the advisor (n=13). Clients expect that their advisor explains his or
her choice of products to them. And while they demand to understand why any
specific product is being recommended (e.g., because of superior performance
or low management fees), some request an even more elaborate reasoning
behind any external product that is offered to them. This would seem in line
with the finding that an open-architecture offering, while commonly
appreciated, disconfirms many clients’ expectations and makes them ‘wonder
why’. One respondent made this very clear: „Well, I would always expect him
[the advisor] to explain to me what this product does, what its features are. And
if it was an external product, then indeed, I’d like to know ‘why this one?’ […]
With one of the bank’s own products, this idea would not occur to me” (male,
30). Note that the order in which the three cues have been presented here is no
indication of their relative importance. Rather, they reflect the natural
“sequence” of the steps that a meeting with a financial advisor would often go
through, as remembered (or imagined) by the interviewee. An advisor would
40
demonstrate “proactiveness” during a first, broader discussion of the options at
hand, the concrete, recommended “product mix” would then indicate whether
competitor products are truly and genuinely a part of the actual offering. Finally,
the advisor’s reasoning behind each product would be “persuasive” or not.
Table 2-3: Typology of salesperson cues 3
Cues
Definitions
Examples
Proactiveness in
offering
open
architecture
The fact that an
advisor either offers
external products on
his or her own
initiative (high
proactiveness) or only
if directly asked to do
so by the client (low
proactiveness)
From my point of view, it would be better if he
(the advisor) brought them (the products) on
right away.
(The advisor should) put them (the third-party
products) on the table at the same time as the
in-house products.
Certainly, a lot is achieved if they (the thirdparty products) are put on the table
proactively...that's a step forward.
I believe that (the proactive offering of thirdparty products) would lower the risk of me
going to see another advisor.
Biased
The client's perception If he tried to foist only bank X products on
product mix that the advisor's
me, I‘d be very skeptical.
recommended choice Well, I‘d question the advice if somebody
either strikes a good
presented only his in-house products to me.
balance between in And when he said ‘I have only this one
house and external
product for you, I can highly recommend it to
products or is strongly
you’, I‘d be very skeptical and I‘d think ‘OK,
biased towards inhe‘ll probably get himself a Porsche from the
house products
Kickbacks.
Only offering external products, that wouldn‘t
be a smart solution. That would be merely a
very ostentatious attempt at saying ‚Hey, I‘m
Mister Neutral‘.
41
Table 2-3: Typology of salesperson cues (cont.) 4
Cues
Definitions
Examples
Persuasiveness of
reasoning
The extent to which
Well, I would always expect him [the advisor]
an advisor's reasoning
to explain to me what this product does, what
behind his or her
its features are. And if it was an external
product recommendproduct, then indeed, I’d like to know ‘why
dation is perceived as
this one?’
credible, convincing
For me, this transparency matters
and transparent
enormously...that I have the feeling these
things are presented to me in a transparent,
forthcoming way.
It‘s obviously much more credible if I (in the
role of the advisor) say ‘Look, there‘s
different options and for each one, these are
the pros and cons‘.
He would have to explain to me why he has
selected exactly those.
He has to tell me why this is good; I want to
know what his reasoning is.
If he says ‘that's what you do today’ or
‘everyone does this’, that's not an argument.
He has to justify his selection and in a way
that I can understand it and say 'that makes
sense.
42
2.4.3.4
Summary
Purpose of the qualitative study was to further explore the relationships
postulated by the high-level conceptual model and to generate more specific
insights into customers’ perception of an open-architecture offering. To
summarize, the results of the qualitative study would seem in line with the basic
relationships forwarded in the previous chapter. Firstly, the findings shed some
initial light on several cues that clients may observe during the advisory episode
and that could subsequently influence their reaction to an offering of third-party
products. Three of these cues seem to bear specific relevance in an openarchitecture context and will therefore receive further attention in the course of
this dissertation: The persuasiveness with which an advisor argues for a specific
choice of products, the proactiveness that an advisor displays in offering thirdparty products and the balance of in-house vs. third-party products in the
advisor’s selection, i.e., the product mix. It should be noted that these cues occur
in different phases of the sales encounter, which suggests that customers observe
and evaluate the advisor’s behavior from start to end. Lastly, the data suggests
that customers respond to an open product architecture with a variety of
attributions. The interview statements include both favorable attributions
regarding the benevolence, expertise or objectiveness of the advisor and
unfavorable ones which suspect ulterior motives on the advisor’s or bank’s side.
While the limited sample size of the qualitative study allows no quantitative
analysis, it should still be noted that customer-oriented attributions were
prevalent. This would support the idea that – due to the general appeal of variety
and/or the disconfirmation of certain long-held beliefs – an open-architecture
offering is perceived favorably by many customers. The fundamental conclusions of Chapter 2 form a concretized conceptual model that is summarized in
Figure 2-3.
43
Figure 2-3: Conceptual model of the relationship between salesperson cues,
customers’ attributions and reactions6
Salesperson cues
Attributions
Customer reactions
Customeroriented
attributions
Persuasiveness
of reasoning
• Satisfaction
• Purchasing
Intention
• Willingness to
provide WOM
Proactiveness in
offering thirdparty products
Balance of
product mix
44
3
HYPOTHESES DEVELOPMENT
In this chapter, a number of hypotheses are developed. They postulate that the
three cues which have been identified – persuasiveness of reasoning, proactiveness and balanced product mix – will affect customer reactions to an openarchitecture offering. These reactions may take form, for instance, in higher
satisfaction, intention to purchase or willingness to recommend. The hypotheses
also point out that the effects of proactiveness and balanced mix on customer
reactions are likely to be moderated by persuasiveness of reasoning. A second
set of hypotheses argues that under certain conditions, these effects are strongly
mediated by customer-oriented attributions.
3.1.1
Impact of Cues on Customer Reactions
3.1.1.1
Persuasiveness of Reasoning
Results of the qualitative study suggest that persuasiveness of reasoning is a
powerful cue. This seems highly plausible since the importance of strong,
convincing messages in sales and service encounters has been highlighted
repeatedly. Research into influencing factors of customer persuasion has demonstrated a positive correlation between argument quality and customers' purchasing intentions (Berger, Cunningham, and Kozinets 1999; Hunt, Smith, and
Kernan 1985; Sanbonmatsu and Kardes 1988). Of specific relevance for the
present study is the contribution of Hunt, Smith and Kernan (1985). The authors
argue that under conditions of expectancy disconfirmation, strong arguments
should have a substantial positive influence on customers’ purchasing intention
and causal attributions, as a more thorough message processing is triggered.
More recent investigations into selling effectiveness have provided further
support by underscoring that meaningful, argument-based communication is a
key skill of successful salespeople (Ahearne, Jelinek, and Jones 2007; Anselmi
45
and Zemanek Jr 1997; Dion and Notarantonio 1992; Plouffe, Hulland, and
Wachner 2009; Rentz et al. 2002) and that a critical element of an effective sales
presentation is for salespeople to "discover relevant needs and, in persuasive
terms, explain corresponding product benefits" (Boorom, Goolsby, and Ramsey
1998, p. 19). For the specific context of personal financial planning services,
Sharma and Patterson (1999) argue that communication effectiveness –
determined by how well a financial advisor communicates to his clients, what
and how much information he shares and how strongly he or she fosters clients'
own understanding – is “the single most powerful determinant of relationship
commitment” (Sharma and Patterson 1999, p. 151). This does not surprise –
financial products are intangible, often complex and require a considerable
explanation effort. Strong sales arguments are often associated with explaining
product features and presenting the advantages and disadvantages of alternative
options (Ahearne et al. 2007; Boorom et al. 1998; Hunt et al. 1985), which
requires salespeople to have considerable expertise on the subject matter. In case
of an open product architecture, clients are very likely to appreciate such expertise: Faced with a seemingly infinite choice of, e.g., investment products,
insurance policies or vacation offers, they may feel the need to rely substantially
on a salesperson’s or advisor’s expert knowledge. Expertise plays a significant
role in building trust (Doney and Cannon 1997; Wood et al. 2008a; Wood et al.
2008b) and hence, it is very likely that a convincing reasoning will lead to
favorable client reactions: It has been shown that a salesperson’s demonstration
of expertise improves customer satisfaction and the overall buyer-seller
relationship (Boles, Johnson, and Barksdale Jr 2000; Crosby et al. 1990).
Satisfaction in turn has often been associated with customers’ loyalty (Anderson
and Sullivan 1993; Bolton and Lemon 1999; Rust and Zahorik 1993) and their
willingness to refer and recommend (Brown et al. 2005; de Matos and Rossi
2008; Rajaobelina and Bergeron 2009; Swan and Oliver 1989).
46
In general, intentions to purchase products or services have been shown to
correlate with the willingness to provide positive word of mouth (Bloemer, de
Ruyter, and Wetzels 1999; Zeithaml 2000). On the basis of these arguments, the
following hypothesis is suggested.
Hypothesis 1: Persuasiveness of reasoning has a significant effect on
Satisfaction, Purchasing intention and WOM Intention
3.1.1.2
Proactiveness
The interview data also seems to indicate that many clients appreciate if the
advisor proactively includes third-party products in his or her product
recommendation. Imagine you wanted to fly to New York and you knew that the
service center agent at SWISS could in theory sell you any airline ticket
available: After you had provided your timing requirements, one of the agent’s
first suggestions would be to book an Emirates flight because it was the option
that best matched your needs. It seems quite likely that such an initiative-taking
would be perceived positively. And even though proactiveness has rarely been
investigated as a behavioral cue that customers distinctly recognize, there is
substantial research that links the proactiveness of salespeople or companies to
their sales and service effectiveness (e.g., Beverland, Farrelly, and Woodhatch
2007; Challagalla, Venkatesh, and Kohli 2009; Worsfold, Worsfold, and
Bradley 2007). Proactive post sales service, for instance, has been shown to lead
to higher customer satisfaction – by preventing misuse and failure of a product
or by soliciting customer feedback (Challagalla et al. 2009). It seems plausible
that in the case of an open-architecture offering, proactiveness my bear an even
greater relevance: First of all, many interviewees in the qualitative study were
unaware that banks offered third-party products in the first place. This seems in
line with reports that at many banks, an open-architecture offering – though
officially promoted – is something that only few clients have yet experienced in
day-to-day reality (Bender 2009; Bolton et al. 2007). Hence, only if the advisor
47
recommends third-party products on his or her own initiative, many clients will
actually learn about such an offering. Secondly, the qualitative study confirmed
that many clients believe a bank will get the highest return on selling its own inhouse products. Thus, it seems to oppose the advisor's own interest if he or she
proactively promotes competitor products. Different lines of marketing and sales
research have shown that consumers often perceive seemingly counter-intuitive
behaviors and messages from salespersons or advertisers as signs of
benevolence, i.e., "an underlying motivation to place the consumer's interest
ahead of self-interest" (Sirdeshmukh et al. 2002, p. 18). The advocacy of an
unexpected position, incongruous or two-sided messaging can improve a
communicator’s
credibility,
as
this
behavior
suggests
sincerity
and
trustworthiness to a client (Crowley and Hoyer 1994; Etgar and Goodwin 1982;
Golden and Alpert 1987; Koeske and Crano 1968; Kohn and Snook 1976;
Pechmann 1992). But can any positive effect of proactiveness prevail even when
the salesperson has given little or poor explanation of why a certain product is
recommended? That seems unlikely, given the need for explanation articulated
by interviewees in the qualitative study. A salesperson’s convincing reasoning
behind his or her product recommendation would seem a ‘table stake’ in most
sales encounters - especially in an advisory context like investment solutions, in
which the explanation of complex, intangible products and their individual
advantages is a key element of the service offered to customers (Sharma and
Patterson 1999). Therefore, it is conceivable that the influence of proactiveness
is moderated by the first cue, persuasiveness of reasoning. In summary, clients
are likely to perceive it as counter-intuitive, benevolent behavior if their advisor
offers them products from other companies, but they might only do so as long as
the advisor acted on his or her own initiative. At the same time, it seems very
likely that a sound reasoning given by the advisor is a “must-have” and without
it, proactiveness may have little positive effect on customer perceptions and
behavioral intentions.
48
These arguments lead to the following hypothesis:
Hypothesis 2: Under high (low) persuasiveness of reasoning, proactiveness in
recommending third-party products has a (has no) significant effect on
Satisfaction, Purchasing intention and WOM Intention
3.1.1.3
Product Mix
Finally, data from the qualitative study suggest that the right balance between
in-house and external products may be another critical cue. The interview data
allow the conclusion that many clients expect the advisor to provide a
‘reasonable’ balance of in-house and third-party products. Consequently, many
clients will find it suspicious if a salesperson is in a position to offer both inhouse and third-party products, but in reality only (or predominately)
recommends in-house ones. And in many cases, such a suspicion would seem
justified: There is well-known risk, for instance, that financial advisors make a
biased product recommendation to their clients because they pursue own
interests (Bolton et al. 2007; Jonas and Frey 2003; Krausz and Paroush 2002). In
fact, most customers assume that a salesperson is biased in one way or another
(e.g. Friestad and Wright 1994; Stafford et al. 1995). Bias has been argued to
take form, for instance, in the withholding of relevant information (Eagly,
Wood, and Chaiken 1978; Jonas and Frey 2003). The availability and adequacy
of other relevant options can certainly be seen as critical information – and this
information is kept from the client if the advisor’s selection is clearly dominated
by products of one provenance. Hence, the very mix of products recommended
by the advisor may indicate a salesperson’s bias and is therefore likely to serve
as a cue to clients. Again, however, it should be asked whether the positive
effect of a well-balanced product mix can prevail when the reasoning behind the
advisor’s selection is unsatisfactory and unconvincing. For reasons similar to
those provided in the previous section, this would seem unlikely. If a
salesperson presented a “fair” mix of different products but ultimately failed to
49
explain in a convincing way why these qualified for recommendation, this
would surely leave most customers unsatisfied. On the other hand, customers
who perceive the salesperson’s individual reasoning as persuasive are likely to
show even more positive reactions when they consider the very mix of
recommended products as unbiased. Hence, the following hypothesis is
forwarded.
Hypothesis 3: Under high (low) persuasiveness of reasoning, the perceived
“balance” of the mix of recommended products has a (has no) significant effect
on Satisfaction, Purchasing intention and WOM Intention
Hypotheses 1,2,3 are illustrated in Figure 3-1.
Figure 3-1: Effects of salesperson behavior on customer reactions7
Advisor Cues
Persuasiveness
of reasoning
Customer Reactions
H1
H2
• Satisfaction
• Purchasing
Proactiveness in
H3
Balance of
product mix
50
Intention
• Willigness to
provide WOM
3.1.2
Causal Attributions as Mediator
The three previous hypotheses have outlined the direct influence that different
cues gained from a salesperson’s behavior may have on customer reactions. The
conceptual model has also forwarded the idea that these effects are likely to be
mediated by attributional thinking. The qualitative pre-study sheds some initial
light on such cognitive processes. The data suggest that customers attribute an
open-architecture offering to a number of different customer-oriented, and less
frequently, suspicion-oriented causes. Given the appeal of variety, and the
positive effect of disconfirmed expectancies, is does not surprise that favorable
attributions prevail when customers form their opinion about an open product
architecture. In the interview sessions, such attributions included assumptions
that the advisor or salesperson was acting in the client's best interest or offered
neutral and objective advice.
Attributions have been shown to substantially determine consumers’ affective
and behavioral responses to an event. Previous research has highlighted a
considerable influence of customers' attributions on their desire to provide wordof-mouth (Bitner 1990; Curren and Folkes 1987; Swanson and Kelley 2001) and
(re-) purchasing intentions (Casado Diaz and Más Ruíz 2002; Folkes et al.
1987). The role that attributions play on customers' affective and behavioral
responses, such as willingness to recommend and purchasing intentions, has also
been confirmed by Hui and Toffoli (2002) and DeCarlo (2005). Recently,
Rajaobelina and Bergeron (2009) have demonstrated for the specific context of
personal financial advisory that perceived customer-orientation has a positive
influence on both purchase intention and word-of-mouth. At the same time,
different studies have argued that attributions are often triggered or influenced
by behavioral cues provided, for instance, by a salesperson (Burnkrant 1975;
DeCarlo 2005; Rose and Dickson 1988; Wood et al. 2008a). In light of both the
antecedents and consequences of attributional thinking, a mediation role seems
to emerge. This notion is supported by DeCarlo (2005), who has explicitly
51
shown how attributions can mediate the effect of a cue – such as salesperson
communication – on consumer attitudes.
3.1.2.1
Mediation of Persuasiveness of Reasoning
Summarizing the above arguments and those that have been forwarded in
Chapter 2, it seems likely that customer reactions instigated by different behavioral cues will be significantly mediated by customer-oriented attributions. If a
salesperson, for instance, delivered a very convincing reasoning for picking a
specific product, a customer may attribute this to the fact that the salesperson
was both understanding and very knowledgeable and consequently would
recommend him or her to others. For the first cue, persuasiveness of reasoning,
this argument is summarized in the following hypothesis:
Hypothesis 4: In an open-architecture sales context, customer-oriented
attributions mediate the influence of reasoning persuasiveness on Satisfaction,
Purchasing intention and WOM Intention
3.1.2.2
Mediation of Proactiveness
The argument that the relationship of cues and customer reactions will be mediated by attributional thinking also applies to the second cue, proactiveness. At
the same time, Hypothesis 2 postulates that the positive effect of proactiveness
on customer reactions is contingent upon the level of persuasiveness. Without a
sound reasoning, proactive behavior has little impact. Hence, any possible mediation of proactiveness can only happen under high persuasiveness – otherwise,
there is no effect to mediate. This argument leads to the following hypothesis:
Hypothesis 5: Under high persuasiveness, customer-oriented attributions
mediate the influence of Proactiveness on Satisfaction, Purchasing intention and
WOM Intention
52
3.1.2.3
Mediation of Product Mix
Finally, a similar reasoning as the above one can be applied to the third cue,
product mix. If a) a balanced product mix leads to favorable customer reactions
only in conjunction with a highly persuasive reasoning and b) such an effect is
likely to be mediated by customer-oriented attributions, then these two
arguments can be summarized in the following hypothesis:
Hypothesis 6: Under high persuasiveness, customer-oriented attributions
mediate the influence of Product Mix on Satisfaction, Purchasing intention and
WOM Intention
Hypotheses 4, 5, 6 are illustrated in Figure 3-2.
Figure 3-2: Mediation through customer-oriented attributions8
Advisor Cues
H4
Persuasiveness
of reasoning
H5
Customeroriented
attributions
Customer
Reactions
H6
H1
Proactiveness in
H2
H3
Balance of
product mix
53
• Satisfaction
• Purchasing
Intention
• Willigness to
provide WOM
4
EXPERIMENTAL ANALYSES
4.1 Overview of Analyses
This chapter presents the design and results of the two experimental studies that
were conducted in order to test the hypothesized effects. Experiments were
chosen as the preferred method for hypothesis testing, as the validation of
theories through the observation of actual service encounters poses a methodlogical challenge. Interactions between individuals can vary substantially and in
combination with different environments, there is an abundance of confounding
variables, or ‘noise’ (Bateson and Hui 1992). Laboratory experiments avoid this
dilemma by offering the researcher high levels of control and by providing a
setting in which individual variables can be manipulated individually and effects
can be isolated to a certain extent. Due to these advantages, experiments are a
common, if not a dominant method in services research.
The independent variables in the focus of this dissertation are different behavioral cues provided by a salesperson. Therefore, a role-playing scenario featuring a salesperson and a client promised to be a highly suitable form of
experimental treatment for the hypothesis testing. Many experiments in psychological and marketing research have employed comparable role-playing
scenarios since they offer control and the experimental stimuli can be manipulated precisely. In addition, previous research indicates that test subjects find
salesperson scenarios to be believable and understandable (e.g. Bitner 1990).
It was feasible and – in order to measure interaction effects – desirable to cover
all cues within one and the same experiment. However, in order to improve the
generalizability of results, it seemed advisable to test the hypotheses in two
different experimental studies. Firstly, it has been argued that the validity of
experimental results is contingent upon the level of realism and authenticity
achieved by the experimental setting (Bateson and Hui 1992). Clearly, one of
the decisive factors in this context must be the type of medium through which
54
the service environment and interaction are created and presented to the test
subjects. One of the key differences between Experiment 1 and 2 was therefore
the type of simulation used to depict the salesperson-customer encounter. The
first experiment featured a written scenario that test subjects were confronted
with at the beginning of the session, an approach commonly used in service
studies (e.g., DeCarlo 2005; Wentzel 2009; White 2005; Wood et al. 2008a).
Stimuli presented through the use of written vignettes offer a very effective
control of the independent variables, but they have been accused of lacking
realism (Bateson and Hui 1992). Therefore, while testing the same hypotheses,
Experiment 2 used a video clip that showed a customer-salesperson interaction
very similar to the written one of Experiment 1. As early as 30 years ago,
researchers have proposed to overcome the disadvantages of text stimuli by
using videotaped interactions as experimental cues: "videotape technology may
offer a particularly appropriate method for isolating the effect of individual
difference variables on attributions of organizational behavior" (Niebuhr, Manz,
and Davis Jr 1981, p. 46). Specifically, it has been argued that video vignettes as
experimental stimuli both offer a high degree of control over the manipulated
variables and confounding effects, and provide greater realism (Grandey et al.
2005). Consequently, several studies report laboratory experiments in which
video stimuli led to results of satisfactory external and internal validity (Baker et
al. 2002; Bateson and Hui 1992; Grandey et al. 2005). Such results would also
seem in line with Bateson and Hui's (1992) finding that videotapes induced
psychological and behavioral effects similar to those in a real-life setting.
A second variation of the setup of Experiment 2 versus Experiment 1 was
achieved by introducing a different line of products and services – the life
insurance business – and by using a sample with different properties. Figure 4-1
provides an overview of the experimental analyses. The design and the results of
the experiments are presented in the following sections.
55
Figure 4-1: Overview of experimental analyses9
EXPERIMENT 1
Independent Variables: Proactiveness, Product mix, Persuasiveness
Stimulus: Written scenario
Participants: 238 Swiss and German banking clients
Hypotheses Tested: 1-6
EXPERIMENT 2
Independent Variables: Proactiveness, Product mix, Persuasiveness
Stimulus: Video scenario
Participants: 260 German adults from an online panel
Hypotheses Tested: 1-6
4.2 Experiment 1
4.2.1 Design, Participants and Procedure
Study 1 employed an online experiment using a 2 (persuasiveness high/low) x 2
(proactiveness high/low) x 2 (product mix balanced/biased) between-subjects
design. A sample of 238 responses was captured from test subjects recruited
through the online community ‘XING’, a business network similar to ‘LinkedIn’
or other online platforms. This approach promised a higher suitability of
subjects for the experiment: The average participant in the sample – German or
Swiss, well into his or her business career and mostly aged between 30 and 50 –
would be more likely to have built own experiences with financial investment
advice than, for instance, students. 53.8% were male, 46.2% were female. In
terms of age, 20.2% were less than 30 years, 72.2% were between the ages of 31
56
and 50 and 7.6% were over the age of 50. The sample distribution across the
eight experimental cells ranged from a low of 25 (10.5%) to a high of 32
(13.4%) participants. All participants received an email including a link to the
online survey homepage. The email informed them also that for each completed
questionnaire, CHF 5 would be donated to a charity project.
Following an introduction on how to proceed in the survey environment, each
participant was randomly assigned to one of 8 written treatments. Similar to
previous research (e.g. Campbell and Kirmani 2000; DeCarlo 2005; Folkes
1984; Swanson and Kelley 2001), the test subjects were instructed to carefully
read the scenario and imagine themselves as the client in the interaction. The
text described a meeting of the participant with a financial advisor at his or her
bank. As stimuli, the scenario comprised a) high/low proactiveness in offering
third-party products, b) biased/balanced mix of selected products and c)
advisor’s persuasive/unconvincing reasoning. The content of all eight scenarios
was identical exclusive of the cue manipulation. After reading the treatment,
participants were asked to respond to several measures of customer-oriented
attributions (e.g., advisor has my best interest at heart) and reactions (e.g.,
intention to purchase). The structure of the experiment is depicted in Figure 4-2.
Figure 4-2: Structure of online experiment 110
General
briefing and
introduction
to the
questionnaire
Exposure to
written
customerclient
interaction
scenario
Measurement
of attributions
57
Measurement
of customer
reactions and
other
variables
The subjects were requested to rate a number of statements, using a Likert scale
with anchors of strongly disagree (1) and strongly agree (7). The different
attributions offered in the questionnaire were based on the results of my
qualitative study. Different methods of measuring attributions have been
discussed by Elig and Frieze (1979): According to their comparisons, intertest
validity and reliability of results seem to be better for structured response
methods than for open-end questions. Consequently, the authors recommend the
use of scale measures, as employed in the present research. A disadvantage of
structured responses may lie in the fact that subjects are prompted with
attributions they might not have formed otherwise (Enzle and Schopflocher
1978). However, such a risk should be reduced by using the specific typology of
attributions which were generated in the preliminary qualitative study. This way,
subjects were confronted only with a selection of attributions that had been
proven to occur in the specific context of an open architecture. Overall, the
rating of predefined attributions would seem common practice in the field of
marketing and sales research (e.g. Dixon, Spiro, and Jamil 2001; Dubinsky,
Skinner, and Whittler 1989; Lichtenstein and Burton 1988).
4.2.2 Manipulation of Independent Variables
Each of the experimental scenarios featured all three cues. In order for the test
subjects to get sufficiently immersed in the experimental scenario, all
participants were exposed to one and the same short introduction text. Its
purpose was to explain the overall situation, introduce the idea of an openarchitecture offering, improve the salience of participants' expectations towards
a bank and its advisor and make them receptive to the specific stimuli tested in
the experiment. The text read as follows:
58
Recently, you've decided to benefit from soaring equity prices on the
German and Swiss stock markets. In order to do this, you've put aside a
certain amount of money that you'd like to invest into suitable mutual
funds. It's important to you that you take a well thought-through, sound
investment decision. Therefore, you have made an appointment with the
bank that you also use for your other financial transactions.
A while ago, you've read that some banks offer a so-called "open product
architecture". Clients of these firms can not only buy in-house funds managed by the bank itself, but they can choose from a wider offering of funds
that includes external products from third parties. You don't know whether
your bank offers such third-party products, but you intend to find out.
You meet with Thomas Breiter, your client advisor, in the lobby of the
bank's local branch. After a short welcome, Mr. Breiter accompanies you to
his office where you take seat at a small conference table. After his assistant has brought you an espresso, you and Mr. Breiter discuss your financial
requirements and expectations for a while. Then, the client advisor suggests
to present you with a selection of products that are suitable for your goals.
4.2.2.1
Manipulation of Proactiveness
‘Advisor’s proactiveness in offering third-party products’ was manipulated by
how actively the advisor brought up the open-architecture offering of his firm
and proposed to include third-party products in his recommendation. The
scenario depicted the advisor as either acting on his own initiative or only on the
client’s explicit demand. The exact wording of the manipulations was as
follows:
High proactiveness: Mr. Breiter rises from his chair, walks over to his PC
and begins to call up a variety of fund profiles. He turns around to you and
says: "Maybe you've heard that our firm has an 'open product architecture'?
As you nod, he continues: "That means you don't get only in-house funds
from us, but also those of other providers. I'd suggest that I'll also include
those third-party products in my selection."
59
Low proactiveness: Mr. Breiter rises from his chair, walks over to his PC
and begins to call up a variety of fund profiles. He turns around to you and
says: "Our in-house funds are really excellent, I'll pick a few for you." You
respond: "Yeah, sure,…but, would it be possible that you also show me a
few funds from other companies?" Mr. Breiter hesitates briefly and
answers: "Ok...sure…if you like…then I can do that."
4.2.2.2
Manipulation of Product Mix
With regard to the manipulation of “product mix”, results from the qualitative
study suggested that in-house products must not strongly dominate the advisor’s
selection if the overall mix is supposed to be perceived as well-balanced. Hence,
the advisor in the scenario proposed six different investment funds out of which
either three or five were from his own bank. The exact manipulations were as
follows:
Balanced product mix: With a focused expression, the client advisor's
glance sweeps across the wide range of different funds that his PC is
displaying. Finally, he prints 1-page profiles for some of the funds and
takes these to the conference table, where he spreads the pages out in front
of you. You leaf through the six documents. Three of the selected funds are
in-house products of the advisor's bank, the other three are from different
external fund managers.
Biased product mix: With a focused expression, the client advisor's glance
sweeps across the wide range of different funds that his PC is displaying.
Finally, he prints 1-page profiles for some of the funds and takes these to
the conference table, where he spreads the pages out in front of you. You
leaf through the six documents. Five of the selected funds are in-house
products of the advisor's bank, only one is from an external fund manager.
60
4.2.2.3
Manipulation of Persuasiveness
‘Persuasiveness of the reasoning behind the advisor’s recommendation’ was
manipulated by how well the advisor managed to explain his reasons for making
a specific recommendation. In one half of the scenarios, he was described as
outlining his reasons in an insightful and transparent way, highlighting pros and
cons. In the other half, he was evasive and used financial jargon instead of sound
arguments. A full description of the treatments can be found in the appendix.
The exact manipulations read as follows:
High persuasiveness: You take a sip of your espresso and give the different
fund profiles another glance. "Ok", you go and lean back in your chair,
"why do you recommend these specific funds?" "Of course", Mr. Breiter
smiles, "let me explain that to you." He then gives you a very accurate and
transparent explanation on why he's chosen each product and what their
individual pros and cons are. He also illustrates very clearly, how these
different funds will offer you a good diversification of your risk.
Low persuasiveness: You take a sip of your espresso and give the different
fund profiles another glance. "Ok", you go and lean back in your chair,
"why do you recommend these specific funds?" "Oh, well" says Mr. Breiter
and shrugs. "It's pretty difficult to explain this in detail". He goes in to longwinded elaboration, using a lot of jargon. Mr. Breiter does not explain the
different pros and cons of each product, but assures that he's got a "good
feeling" about the selection he has recommended.
4.2.3 Selection of Measures
4.2.3.1
Dependent Measures
In Experiment 1, four dependent variables were measured: customer-oriented
attributions, satisfaction, purchasing intention and willingness to provide wordof-mouth (WOM). All variables were measured as multi-item constructs,
61
leveraging existing scales whenever possible. All items had been pretested for
face validity, asking 20 clients of different banks for a check on clarity and
unambiguousness. Participants rated their customer-oriented attributions on five
seven-point scales (applies very much / does not apply at all) which had been
created based on the results of the qualitative pre-study. The reliability level of
the scale (Į = .83) was satisfactory. Table 4-1 provides a list of the items that
comprised the customer-oriented attributions measure.
Table 4-1: Attribution measurement items5
Please indicate your agreement with the following statements on a scale
from "do not agree at all" to "completely agree"
Mr. Breiter is interested in my welfare and long-term benefit
Mr. Breiter is very motivated to achieve my goals
Mr. Breiter is a neutral advisor
The advice that Mr. Breiter gives is objective
Mr. Breiter's recommendations are unaffected by him being
employed by a bank
With regard to the different customer reactions, satisfaction was measured using
three items (very satisfied / not at all satisfied with the service, like very
much/do not like at all what has been done, has highly fulfilled / not at all
fulfilled my requirements) that Hui et al (2004, referring to Westbrook 1980)
have applied. In order to measure purchasing intention, a 3-item scale was taken
from De Carlo (2005) and slightly adapted to fit a service encounter in which
investment funds are discussed (very likely/very unlikely to use Mr. Breiter’s
support, very likely/unlikely to invest at Mr. Breiter’s bank, very likely/very
unlikely to purchase products from Mr. Breiter). Finally, the intention to provide
Word-of-Mouth was measured with a 3-item scale (very likely / very unlikely to
provide WOM, very likely / very unlikely to recommend, very likely / very
unlikely to suggest to friends) taken from Maxham III and Netemeyer (2002).
All items employed seven-point Likert scales.
62
4.2.3.2
Manipulation Checks
In order to ensure that the three different salesperson cues were manipulated
successfully, participants of Experiment 1 rated each stimulus on two sevenpoint scales. In each case, the second measurement scale was reversed to counter
response-set effects. Participants rated two statements, assessing the persuasiveness of the advisor (“Mr. Breiter made an effort to explain his recommendations
well”, “The reasoning Mr. Breiter gave for his fund selection was hardly convincing”). The two items yielded a satisfactory reliability level of r = .70.
Table 4-2: Overview of measures used in experiment 16
Number
of Items
Reliability Source
Customer-oriented attrib.
Satisfaction
Intention to purchase
Willingness to provide WOM
5
3
3
3
α = .83
α = .96
α = .96
α = .98
Manipulation Check
Persuasiveness
Proactiveness
Product mix
2
2
2
r = .70
r = .87
r = .82
Measure
Dependent Variables
Hui et al (2004)
DeCarlo (2005)
Maxham III and
Netemeyer (2002)
For proactiveness, the statements to rate were “Mr. Breiter has offered me thirdparty funds on his own initiative” and “only at my request, Mr. Breiter has
offered me funds that were not from his own bank”. The level of reliability (r =
.87) was good. Finally, in order to measure the manipulation of product mix, the
participants assessed the statements “the advisor has offered me a balanced mix
63
of in-house and external products” and “products of his own bank dominated
the advisor’s product selection” (r = .82). All items were anchored by applies
very much / does not apply at all. The different dependent variables and
manipulation checks and their reliability measures are listed in Table 4-2.
4.2.4 Results
4.2.4.1
Manipulation Checks and Item Reliability.
Firstly, a series of manipulation checks was applied to ensure that the different
manipulations had been effective. The salesperson’s reasoning was considered
more persuasive under conditions of high persuasiveness than under low persuasiveness (MHighPers = 4.56, MLowPers = 1.83, F(1,238) = 275.104, p < .001). While
this intended effect was by far the strongest, marginally significant effects were
also measured for the independent variable proactiveness (F(1,238) = 2.87, p <
.093) and for the interactive effects of persuasiveness/proactiveness (F(1,238) =
3.17, p < .077), persuasiveness/product mix (F(1,238) = 2.90, p < .091) and
proactiveness/product mix (F(1,238) = 3.12, p < .08). Product mix by itself
(F(1,238) = 1.64, p > .20) and the interaction of all three independent variables
(F(1,238) = .93, p > .33) had no significant effect on the perception of persuasiveness. In conclusion, the manipulation of the independent variable persuasiveness was effective, but not as precise as could be desired.
Respondents who had received a scenario treatment with high proactiveness reported a greater value for the advisor’s proactiveness than in those in the low
proactiveness condition (MHighProact = 5.17, MLowProact, = 1.51, F(1,238) = 416.73,
p < .001). Furthermore, it was tested whether the other independent variables,
persuasiveness or product mix, or any interactions among the independent variables had an effect on the manipulation check. The main effects of neither persuasiveness (F(1,238) = 1.09, p > .29) nor product mix (F(1,238) = 1.36, p > .24)
were significant. The same applies to the interaction effects of persuasive-
64
ness/proactiveness (F(1,238) = .22, p > .64), persuasiveness/product mix
(F(1,238) = .01, p > .91), proactiveness/product mix (F(1,238) = 2.49, p >.11)
and the interaction of all three independent variables (F(1,238) = .54, p > .46).
Hence, it is concluded that the manipulation of proactiveness was successful.
A third manipulation check revealed that respondents in the balanced-mix condition reported a considerably higher value for balanced product mix than those
in the biased product mix condition (MBalancedMix = 4.08, MBiasedMix = 1.62,
F(1,238) = 209.22, p < .001). However, this manipulation would seem to have
been the least accurate one: significant effects were also measured for the independent variables persuasiveness (F(1,238) = 3.15, p < .078) and proactiveness
(F(1,238) = 27.18, p < .001), as well as for the interactive effects of product mix
and proactiveness (F(1,238) = 11.63, p < .002). The data suggest no significant
effect of the interactions of persuasiveness/product mix (F(1,238) = .12, p >
.73), persuasiveness/proactiveness (F(1,238) = .02, p > .89) and the interaction
of all three independent variables (F(1,238) = .67, p > .41).
Reliability was assessed using the Cronbach’s alpha coefficient, which yielded
satisfactory values for the construct of 'customer-oriented attributions' (0.83), for
satisfaction (0.96), purchasing intention (0.96) and willingness to provide WOM
(0.98). These values can also be found in Table 4-2.
4.2.4.2
Hypothesis Testing
Effect of persuasiveness on customer reactions. A three-factor multivariate
analysis of variance (MANOVA) was used to test for the main and interactive
effects of cues on customer reactions. Hypothesis 1 predicted that
persuasiveness would have a significant effect on customer reactions. The
results of the first experiment (Table 4-3) show that persuasiveness of reasoning
indeed had substantial influence on customers’ satisfaction (Fsatis(1,238) = 80.96,
p < .001). their intention to purchase (Fpurchase(1,238) = 44.99, p < .001), and
their willingness to provide word-of-mouth (FWOM(1,238) = 55.52, p < .001).
65
as dependent variable, main effects 7
Dependent Variable
F(1,238)
p
Persuasiveness
Satisfaction
80.96
p <. 001
of Reasoning
44.99
p <. 001
(high/low)
55.52
p <. 001
Proactiveness
Satisfaction
20.46
p <. 001
(high/low)
13.63
p <. 001
16.64
p <. 001
Product Mix
Satisfaction
13.97
p <. 001
(balanced /
11.37
p <. 002
biased)
8.35
p <. 005
experiment 1, main effects 8
Low PersuasiveHigh Persuasiveness of Reasoning ness of Reasoning
Mean
Mean
Dependent Variables
Satisfaction
1,83
0,96
2,09
1,19
1,47
0,87
Manipulation Check
1,83
1,11
Note: Numbers in italic letters are standard deviations
3,25
1,57
3,29
1,58
2,64
1,55
4,56
1,42
Respondents were more satisfied when persuasiveness was high (Msatis = 3.25)
than when it was low (Msatis = 1.83). As Table 4-4 shows, they were also more
intent to purchase (high persuasiveness: Mpurchase = 3.29, low persuasiveness:
66
Mpurchase = 2.09) and more willing to provide word-of-mouth (high persuasiveness: MWOM = 2.64, low persuasiveness: MWOM = 1.47). This supports H1.
Moderation of proactiveness. In Hypothesis 2, it was postulated that proactiveness would only have a positive effect on customer reactions under conditions of
high persuasiveness of reasoning. The experimental results show such a
significant interaction for all three variables (Fsatis(1,238) = 21.44, p < .001;
Fpurchase(1,238) = 14.37, p < .001; FWOM(1,238) = 23.58, p < .001). The
corresponding values are also listed in Table 4-5. As Figure 4-3 shows, customer
reactions were only improved through high proactiveness when persuasiveness
was high (high proactiveness: Msatis=3.88; Mpurchase=3.86; MWOM=3.25; low
proactiveness: Msatis =2.55; Mpurchase =2.65; MWOM =1.96). Under conditions of
low persuasiveness, proactiveness had no significant effect (high proactiveness:
Msatis=1.82; Mpurchase=2.08; MWOM=1.42; low proactiveness: Msatis =1.84; Mpurchase
=2.10; MWOM =1.53). All values are listed in Table 4-6.
Table 4-5: Results of multivariate analyses in experiment 1; customer
reactions as dependent variables, interaction effects 9
Dependent Variable
F(1, 238)
p
Persuasiveness x
Proactiveness
Satisfaction
21.44
14.37
23.58
p < .001
p < .001
p < .001
Persuasiveness x
Product Mix
Satisfaction
9.03
0.88
4.06
p < .004
p > .34
p < .05
For high persuasiveness, a t-test of contrast effects yielded two-tailed p-values <
.001 for all three customer reaction variables. For low persuasiveness, no
significant effects were measured, with p-values > 0.9 (satisfaction), > 0.9
(intention to purchase) and > 0.45 (willingness to provide WOM). These
findings are fully consistent with H2.
67
experiment 1, interaction persuasiveness x proactiveness 10
Low Persuasiveness
of Reasoning
High Persuasiveness of
Reasoning
Low Pro- High Pro- Low Pro- High Proactiveness activeness activeness activeness
Dependent Variables
Satisfaction
Manipulation Check
1,84
0,94
2,10
1,21
1,53
0,87
1,82
1,00
2,08
1,18
1,42
0,87
2,55
1,25
2,65
1,25
1,96
1,14
3,88
1,58
3,86
1,63
3,25
1,63
1,46
1,05
5,02
1,78
1,56
0,88
5,31
1,58
customer reactions as dependent variables11
4,5
Satisfaction
4,0
3,5
3,0
Proactiveness
LOW
2,5
2,0
Proactiveness
HIGH
1,5
1,0
0,5
0,0
Persuasiveness Persuasiveness
LOW
HIGH
68
Figure 4-3 (cont.): Interaction of persuasiveness and proactiveness in
experiment 1; customer reactions as dependent variables12
4,5
4,0
Purchasing
Intention
3,5
3,0
2,5
2,0
1,5
Proactiveness
LOW
Proactiveness
HIGH
1,0
0,5
0,0
LOW
HIGH
4,5
4,0
3,5
3,0
2,5
2,0
1,5
1,0
0,5
0,0
Willingness to
provide WOM
Proactiveness
LOW
Proactiveness
HIGH
LOW
HIGH
69
Moderation of product mix. Hypothesis 3 postulated that product mix would
have a positive effect on customer reactions only when persuasiveness of
reasoning was high. The data provide evidence of such an effect for the
dependent variables satisfaction and willingness to provide WOM (Fsatis(1,238)
= 9.03, p < .004; FWOM(1,238) = 4.06, p < .05). For purchasing intention, no
such moderating influence was detected (Fpurchase(1,238) = 0.88, p > .34).
Instead, it would seem that product mix always has a significant effect on
purchasing intention, independent of the persuasiveness of arguments. The
results are listed in Table 4-5. As the mean values in Table 4-7 demonstrate,
satisfaction , purchasing intention and willingness to provide WOM were
significantly higher under conditions of high persuasiveness and balanced
product mix (Msatis =3.70; Mpurchase =3.59; MWOM =2.95) than under high
persuasiveness and biased product mix (Msatis =2.76; Mpurchase =2.96; MWOM
=2.31). No such effect could be found when persuasiveness was low (Balanced
product mix: Msatis =1.89; MWOM =1.54; biased product mix: Msatis =1.77; MWOM
=1.41), with the exception of purchasing intention (balanced product mix:
Mpurchase =2.30; biased product mix: Mpurchase =1.89). The interaction effects are
illustrated in Figure 4-4.
0.05 for all three customer reaction variables. For low persuasiveness, p-values
were > 0.5 (satisfaction), and > 0.4 (willingness to provide WOM). In line with
the above MANOVA results, the p-value for intention to purchase was
significant at < .06. With the exception of the variable purchasing intention, the
results support Hypothesis 3.
70
experiment 1, interaction persuasiveness x product mix 11
Dependent Variables
Satisfaction
Manipulation Check
Low Persuasiveness
of Reasoning
High Persuasiveness
of Reasoning
Biased
Balanced
Prod.Mix Prod.Mix
Biased
Balanced
Prod.Mix Prod.Mix
1,77
1,01
1,89
1,14
1,41
0,86
1,89
0,91
2,30
1,22
1,54
0,89
2,76
1,41
2,96
1,54
2,31
1,42
3,70
1,60
3,59
1,57
2,95
1,62
1,49
0,77
3,90
1,77
1,76
1,09
4,25
1,74
customer reactions as dependent variables 13
4,0
Satisfaction
3,5
3,0
2,5
2,0
Product Mix
BIASED
1,5
1,0
Product Mix
BALANCED
0,5
0,0
LOW
HIGH
71
4,0
3,5
Purchasing
Intention
3,0
2,5
2,0
Product Mix
BIASED
1,5
1,0
Product Mix
BALANCED
0,5
0,0
LOW
HIGH
3,5
Willingness to
provide WOM
3,0
2,5
2,0
Product Mix
BIASED
1,5
1,0
Product Mix
BALANCED
0,5
0,0
LOW
HIGH
After the first section of the analysis had investigated a) the direct relationships
between cues and customer reactions and b) the interaction among cues, the
second part focused on the mediating role of customer-oriented attributions. In
order to validate the corresponding hypotheses 4, 5 and 6, initial tests addressed
72
the direct influence of the three cues on customer-oriented attributions and,
again, potential interaction effects. Subsequent testing then incorporated a
stepwise analysis of all three sets of variables – cues, attributions and reactions.
Effect of persuasiveness on customer-oriented attributions. The conceptual
model outlined in Chapter 2 postulates that the three behavioral cues not only
will influence customer reactions, but are also likely to trigger attributional thinking. A three-factor unvariate analysis of variance (ANOVA) was therefore used
to investigate whether main and interaction effects comparable to those found
with customer reactions also occurred with ‘customer-oriented attributions’ as
the dependent variable. The results of the first experiment (Table 4-8) show that
persuasiveness of reasoning indeed had substantial influence on customeroriented attributions (Fattr(1,238) = 20.93, p < .001): Respondents were more
likely to attribute favorable motives behind the advisor’s actions when persuasiveness was high (MHI Persuas = 2.78) than when it was low (MLO Persuas = 2.25).
Table 4-8: Results of a univariate analysis in experiment 1; customeroriented attributions as dependent variables12
F(1, 238)
p
Persuasiveness (High / Low)
Proactiveness (High / Low)
Product mix (Balanced / Biased)
20.93
13.06
27.04
p < .001
p < .001
p < .001
Persuasiveness x Proactiveness
Persuasiveness x Product Mix
8.79
0.03
p < .004
p > .85
Moderation of proactiveness and product mix. Yet again, the data suggests a
significant interaction between persuasiveness of reasoning and proactiveness
(Fattr(1,238) = 8.79, p < .004). Under high persuasiveness, more favorable attributions were triggered when proactiveness was high than when it was low (MHI
73
Proact
= 3.16; MLO
Proact
= 2.44). When persuasiveness was low, no such effect
occurred (MHI Proact = 2.27; MLO Proact = 2.20). Planned contrasts yielded values of
t(118) = -3.75, p < .001) for high persuasiveness, and t(120) = -0.43, p > .66)
for low persuasiveness.
In the case of product mix, no such moderation could be demonstrated: Product
mix always had a significant influence on customer oriented attributions, independent of the level of persuasiveness (Fattr(1,238) = 0.03, p > .85). Mean values
were: high persuasiveness, MBalanced Mix = 3.07; MBiased Mix = 2.54; low persuasiveness, MBalanced Mix = 2.56; MBiased Mix = 1.93. Mean values for all three effects can
also be found in table 4-9.
A t-test delivered values of t(118) = -2.66, p < .01 for high persuasiveness and
t(120) = -4.48, p < .001 for low persuasiveness. The interactions are illustrated
in Figure 4-5.
variable in experiment 113
Dependent Variables
Low Persuasiveness
2,24
0,83
High Persuasiveness
2,82
1,11
Low Persuasiveness
High Persuasiveness
Low Proactiveness
2,20
0,84
Low Proactiveness
2,44
1,00
High Proactiveness
2,27
0,83
High Proactiveness
3,16
1,09
Low Persuasiveness
High Persuasiveness
Biased
Prod.Mix
1,93
0,68
Biased
Prod.Mix
2,54
1,02
Balanced
Prod.Mix
2,56
0,85
74
Balanced
Prod.Mix
3,07
1,13
Figure 4-5: Interaction of persuasiveness and proactiveness / product mix;
customer-oriented attributions as dependent variable 15
3,5
3,0
Proactiveness
2,5
2,0
Proactiveness
LOW
1,5
1,0
Proactiveness
HIGH
0,5
0,0
LOW
HIGH
3,5
Product Mix
3,0
2,5
2,0
Product Mix
BIASED
1,5
1,0
Product Mix
BALANCED
0,5
0,0
LOW
HIGH
Mediation of persuasiveness through customer-oriented attributions. The results
of Experiment 1 suggest a significant main effect of persuasiveness on both customer reactions and customer-oriented attributions. At the same time, it has been
75
hypothesized that customer-oriented attributions are likely to mediate the influence of cues on customer reactions. In order to examine if customer-oriented
attributions mediate the impact of persuasiveness, a further analysis was undertaken. The procedure followed the recommendations from Baron and Kenny
(1986) who apply a sequence of regression analyses to test for such effects.
Firstly, persuasiveness (dummy variable: low=0, high=1) had an impact on the
dependent variables (ȕsatisf = .48, p < .001; ȕpurchase = .40, p < .001; ȕWOM = .42, p
<.001). Secondly, persuasiveness was also related to customer-oriented
attributions (ȕattr = .29, p < .001). Thirdly, customer-oriented attributions were a
significant predictor of the dependent variables (ȕsatisf = .66, p < .001; ȕpurchase =
.60, p < .001; ȕWOM = .68, p <.001). Lastly, when both persuasiveness and
customer-oriented attributions were included in the regression model, the
mediator remained a significant predictor (ȕsatisf = .57, p < .001; ȕpurchase = .53, p
< .001; ȕWOM = .61, p <.001), whereas the impact of the independent variable
decreased (ȕsatisf = .32, p < .001; Sobel: z = 4.27; ȕpurchase = .24, p < .001; Sobel: z
= 4.18; ȕWOM = .25, p <.001; Sobel: z = 4.31). Thus, H4 is confirmed. The results
are summarized in Figure 4-6.
attributions in experiment 116
0.29
Customeroriented
attributions
Persuasiveness
of reasoning
Satisfaction: 0.57 (0.66)
Puchasing Int.: 0.53 (0.60)
WOM: 0.61 (0.68)
Customer
reactions
WOM: 0.25 (0.42)
Note: The total effect between the predictor and the criterion (i.e., before controlling for the
mediator) is given in parentheses; the direct effect (i.e., after controlling for the mediator) is
given outside the parentheses. p was significant at < .001 level
76
Mediation of proactiveness through customer-oriented attributions. The above
results suggest that the influence of persuasiveness on customer reactions is
substantially mediated by customer-oriented attributions. The following section
investigates whether a similar effect occurs for the influence of proactiveness
and product mix. Previously, it has been demonstrated that proactiveness and
product mix only lead to more favorable customer reactions when persuasiveness is high. Hence, if the influence of proactiveness and product mix should
also be mediated by customer-oriented attributions, this can only be the case
under conditions of high persuasiveness – otherwise, there is no significant
effect to be mediated. In order to validate Hypotheses 5 and 6, two separate
analyses were performed, splitting the sample into high (n=117) and low
persuasiveness (n=119) groups. Then, the same approach that had been used to
test for a mediation of persuasiveness was used on proactiveness. Firstly,
proactiveness (dummy variable: low=0, high=1) had an impact on the dependent
variables when persuasiveness was high (ȕsatisf = .42, p < .001; ȕpurchase = .39, p <
.001; ȕWOM = .42, p <.001), but had no such effect under low persuasiveness
(ȕsatisf = -0.01, p > 0.9; ȕpurchase = -0.01, p > 0.9; ȕWOM = -0.07, p > 0.4). It should
be noted that only under conditions of high persuasiveness, proactiveness was
also related to customer-oriented attributions, a prerequisite for any mediation
through this variable (low persuasiveness: ȕattr = 0.04, p > .60; high persuasiveness: ȕattr=.33, p < .001). Thirdly, customer-oriented attributions were always a
significant predictor of the dependent variables (low persuasiveness: ȕsatisf=.54, p
< .001; ȕpurchase=.51, p < .001; ȕWOM=.49, p <.001; high persuasiveness:
ȕsatisf=.66, p < .001; ȕpurchase=.59, p < .001; ȕWOM=.72, p <.001).
Lastly, when both proactiveness and customer-oriented attributions were included in the regression model, the mediator remained a significant predictor
(low persuasiveness: ȕsatisf = .54, p < .001; ȕpurchase = .51, p < .001; ȕWOM = .50, p
< .001; high persuasiveness: ȕsatisf = .58, p < .001; ȕpurchase = .52, p < .001; ȕWOM =
.65, p < .001), whereas the impact of the independent variable was diminished
(low persuasiveness: ȕsatisf = -0.03, p > .7; Sobel: z = .43; ȕpurchase= -0.03, p > .70;
77
Sobel: z = .43; ȕWOM=. -0.09, p > .25; Sobel: z = .43; high persuasiveness: ȕsatisf
= .23, p < .002; Sobel: z = 3.42; ȕpurchase = .22, p < .007; Sobel: z = 3.28; ȕWOM =
.20, p <.005; Sobel: z = 3.52). The results are illustrated in Figure 4-7.
Consistent with H5, it can be concluded that under high persuasiveness, the
effect of proactiveness on customer reactions is mediated by customer-oriented
attributions.
Figure 4-7: Mediation of proactiveness through customer-oriented attribution
( under conditions of high persuasiveness) in experiment 1 17
0.33
Customeroriented
attributions
WOM: 0.65 (0.72)
Customer
reactions
Proactiveness
Satisfaction: 0.23* (0.42)
Puchasing Int.: 0.22** (0.39)
WOM: 0.20* (0.42)
given outside the parentheses. p was significant at < .001 level., < .005 (*) or .01 (**)
Mediation of product mix through customer-oriented attributions. A final
analysis examined if customer-oriented attributions also mediated the impact of
product mix under high persuasiveness conditions. Again, it was necessary to
perform separate analyses for the two persuasiveness conditions since the
relationship between product mix and customer reactions should be positive in
the high persuasiveness and insignificant in the low persuasiveness conditions.
Firstly, product mix (dummy variable: low=0, high=1) had an impact on the
dependent variables (low persuasiveness: ȕsatisf = .06, p > .50; ȕpurchase = .17, p <
78
.06; ȕWOM = .08, p > .40; high persuasiveness: ȕsatisf = .30, p < .002; ȕpurchase = .20,
p < .03; ȕWOM = .21, p <.03). This influence only occurred under high
persuasiveness, with the exception of purchasing intention, on which product
mix always has an effect. Both findings are in line with previous evidence
discussed in the context of Hypothesis 3. Secondly, product mix was also related
to customer-oriented attributions (low persuasiveness: ȕattr =.38, p <.001; high
persuasiveness: ȕattr=.24, p < .01). In congruence with earlier ANOVA results,
the significance of this relationship was not contingent upon the level of
persuasiveness. Thirdly, customer-oriented attributions were always a
significant predictor of the dependent variables (low persuasiveness: ȕsatisf = .54,
p < .001; ȕpurchase = .51, p < .001; ȕWOM = .49, p <.001; high persuasiveness: ȕsatisf
= .66, p < .001; ȕpurchase = .59, p < .001; ȕWOM = .72, p <.001). Lastly, when both
product mix and customer-oriented attributions were included in the regression
model, the mediator remained a significant predictor (low persuasiveness:
ȕsatisf=.61, p < .001; ȕpurchase=.52, p < .001; ȕWOM=.54, p <.001; high persuasiveness: ȕsatisf = .62, p < .001; ȕpurchase = .57, p < .001; ȕWOM = .71, p <.001), whereas
the impact of the independent variable decreased (low persuasiveness: ȕsatisf = 0.17, p < .04; Sobel: z = 3.81; high persuasiveness: ȕsatisf =.15, p < .04; Sobel: z
= 2.54) or was eliminated (low persuasiveness: ȕpurchase = -.02, p > .70; Sobel: z =
3.59; ȕWOM= -.13, p > .12; Sobel: z = 3.66; High persuasiveness: ȕpurchase = .06, p
> .40; Sobel: z = 2.49; ȕWOM = .04, p >.50; Sobel: z = 2.58). The results show
that under conditions of high persuasiveness, the effect of product mix on
customer reactions is partially (satisfaction) or fully (purchasing intention and
WOM willingness) mediated by customer-oriented attributions. Thus, H6 is
confirmed.
79
(under conditions of high persuasiveness) in experiment 118
0.24**
Customeroriented
attributions
WOM: 0.71 (0.72)
Customer
reactions
Product mix
Puchasing Int.: n.s.*** (0.20)*
WOM: n.s. *** (0.21)*
given outside the parentheses. p was significant at < .002 level., < .04 (*) or .01 (**).
Insignificant p was > .40 (***)
4.2.5 Discussion
Study 1 indicated that in an open-architecture sales context, the persuasiveness
of reasoning, the proactiveness in offering third-party products and the mix of
recommended products are powerful cues that are perceived by clients and
affect their reactions. While persuasiveness of reasoning has a significant main
effect on customer reactions, it also moderates the influence of the other two
cues. Proactiveness and product mix only had a positive effect on customer
reactions when at the same time persuasiveness was high. One exception must
be noted for the dependent variable purchasing intention, for which no
significant interaction between persuasiveness and product mix could be found.
Of the three customer reaction variables, the intention to purchase would seem
to have the most immediate link to the investment decisions and risks that a
client ultimately takes. Hence, it seems plausible that with regard to the actual
purchasing intention, a “balanced” product mix – and the risk diversification it
80
may promise – is of substantial relevance to a client, independent of other
factors, such as a convincing reasoning of the advisor.
The results from Study 1 also provide first empirical evidence of the mediating
function that attributions seem to exert. The results suggest a substantial
influence of all three cues on customer-oriented attributions, which in turn have
a strong effect on the different customer reactions. The effect of persuasiveness
on customer reactions shrinks significantly upon the addition of customeroriented attributions as mediator. For proactiveness and product mix, this effect
is contingent on the level of persuasiveness and can achieve full mediation in
some cases. Taken together, Experiment 1 provides satisfactory evidence to
support all six hypotheses.
4.3 Experiment 2
4.3.1 Design, Participants and Procedure
As in the previous research, Experiment 2 employed an online, scenario-based
setup using a 2 (persuasiveness high/low) x 2 (proactiveness high/low) x 2
(product mix balanced/biased) factorial design. With the help of a professional
market research agency, a sample of 260 responses was gathered from participants recruited through an online panel representative of the German population.
49.2% of the test subjects were male, 50.8% were female. 32% were less than 30
years old, 46.6% were between the ages of 30 and 50 and 21.4% were over the
age of 50. The relative size of the age groups illustrates that the sample was
remarkably different from the one used in Study 1. The sample distribution
across the eight experimental cells ranged from a low of 29 (11.2%) to a high of
39 (15%) participants. All participants received an email including a link to the
online survey homepage. They were also informed that the completion of the
questionnaire would be rewarded with an amount of 2.50 EUR.
81
4.3.2 Manipulation of Independent Variables
Manipulation of the variables was similar to Experiment 1. This time, however,
the treatment was presented in form of a video clip embedded into the online
questionnaire. As stimulus material, video clips in 8 different combinations of
high/low persuasiveness, high/low proactiveness and balanced/biased product
mix were produced. Filming of the sales encounters took place in the client
meeting room of a local highstreet bank, an environment sufficiently similar to
the branch of an insurance company. Adding to the realism, the insurance agent
wore a business suit and tie; different insurance brochures were spread out on a
nearby sideboard. The premise of the video was that a customer had visited an
insurance agent and they had talked about the customer’s intention to take out a
life insurance policy. He was now sitting in the agent’s office waiting for the
agent to recommend a number of options. Similar to the approach taken in
Experiment 1, all participants were exposed to a short introduction text in order
to prepare the test subjects for the experimental scenario. The text explained the
situation and introduced the idea of an open-architecture offering to the
participants. A main objective was to make participants' expectations salient and
to improve their receptiveness to the specific stimuli of the experiment. The text
read as follows:
Mr. Staiger has decided to put some money aside by taking out a life
insurance. It's important to him to choose the insurance product most
appropriate for his needs. Therefore, he's made an appointment with the
local agency of a large insurance company. A while ago, Mr. Staiger has
read that some insurance firms offer a so-called "open product
architecture". Their customers can buy the firm's own-label insurance
policies, but also those of third-party providers. He intends to find out
whether his insurance company will also offer such third-party products.
After Mr. Müller, the insurance agent, has welcomed him at his office, the
two of them discuss Mr. Staiger's financial requirements and expectations
for a while. Then, Mr. Müller suggests to present his customer with a
selection of insurance products that will meet his requirements.
82
Both the customer and agent were male and approximately thirty-five years of
age. The customer did not vary his reactions across the video clips but the agent
did in order to simulate the eight experimental conditions. The video mainly
showed the insurance agent from the waist up, talking to the customer, or
standing in front of a sideboard, sifting through different insurance brochures. In
order to minimize the influence of the on-screen customer on the participants,
his script was limited to few vocal cues, and his tone of voice was kept
sufficiently neutral. The video scenarios were enacted and re-filmed a number of
times until both actors’ behavior was consistent and realistic within each scene.
The content of the eight clips was identical exclusive of the cue manipulation.
Each of the eight movies consists of three scenes. The first scene starts out
identically for both conditions: Mr. Müller, the insurance agent, concludes a
conversation he has had with his client before. He suggests selecting a few
product alternatives that will suit the client's requirements. After a cut, the clip
shows the agent standing in front of a sideboard, browsing through different
product brochures.
Figure 4-9: Screenshot of the experimental video clip19
83
In condition one, high proactiveness, Mr. Müller turns around, facing the
customer, and mentions that his firm has an open-architecture offering and
explains that this means the customer can also purchase third-party products.
The agent suggests including some of these in his recommendation. In condition
two – low proactiveness – Mr. Müller turns around, pointing out that his firm
has "great" in-house products and that he's picking a few of those. On that
remark, the customer, Mr. Staiger, nods his approval, but asks explicitly whether
the agent can also include some third-party products in his selection. Mr. Müller
seems surprised, but after a short hesitation agrees to satisfy this request. In
scene two, the sales agent sits down at the table again and spreads out six
brochures in front of his client. Under the condition of balanced product mix, he
explains that three of the leaflets are about in-house products, while the other
three promote policies from other insurance companies. Under the condition of
biased product mix, five out of six brochures are in-house material and the
advisor just points his finger at a sixth one which is for a competitor product.
Finally, in scene three, the customer asks the insurance agent to explain in more
detail why he has picked the specific products which are on the table. In
condition one, high persuasiveness of reasoning, the agent provides an elaborate
and clear explanation of some of the key differences among the products. In
condition two, the agent seems evasive, argues that the strategy behind his pick
would be "difficult to explain" and uses a lot of jargon. He does not explain the
different pros and cons of each product, but assures that he's got a "good
feeling" about the selection he has recommended.
Following an introduction on how to proceed in the online survey environment,
each participant was randomly assigned to one of the eight video treatments.
The survey questionnaire was identical to the one used in Experiment 1, with
only few adaptations to the different industry (i.e., “insurance policies” instead
of “investment funds”, “insurance agent” instead of “advisor”). After watching
the video clip, participants were asked to respond to several measures of
attributions, perceptions and behavioral intentions (see Figure 4-10).
84
Figure 4-10: Structure of online experiment 220
General
information
and handling
of the
questionnaire
Exposure to
video clip of a
customerclient
interaction
Measurement
of attributions
Measurement
of customer
reactions and
other
variables
4.3.3 Selection of Measures
4.3.3.1
Dependent Measures
In order to ensure comparability of results, Experiment 2 employed the same
measures for all dependent variables as Experiment 1, listed in Table 4-10.
Customer-oriented attributions were measured with five items, satisfaction,
purchasing intention and willingness to provide WOM were measured with three
items each. Their wording had to be slightly adapted to the different
experimental setup: While in the written scenario of Experiment 1, the
participant had been addressed as the client, he or she observed an interaction
between a client and a salesperson in Experiment 2. Consequently, the
participants were asked to evaluate the scenario as if this episode had happened
to them. The individual measurement items were therefore phrased in
subjunctive form (e.g., “I would be satisfied with this service” rather than “I am
satisfied”).
4.3.3.2
Manipulation Checks
As previously, the manipulation checks each featured two seven-point scales
rated by participants. The two items used for assessing the persuasiveness of the
insurance agent (“Mr. Müller made an effort to explain his recommendations
well”, “The reasoning Mr. Müller gave for his policy selection was hardly
85
convincing.”) yielded a satisfactory reliability level of r = .72. For proactiveness, the statements to rate were “Mr. Müller has offered me third-party policies
on his own initiative” and “only at my request, Mr. Müller has offered me
policies that were not from his own insurance firm”. The level of reliability (r =
.89) was good. Finally, in order to measure the manipulation of product mix, the
participants assessed the statements “the agent has offered me a balanced mix of
in-house and external products” and “products of his own insurance company
dominated the agent’s product selection” (r = .75). All items were anchored by
applies very much / does not apply at all.
The different dependent variables and manipulation checks and their reliability
measures are listed in Table 4-9.
Table 4-10: Overview of measures used in experiment 214
Measure
Number
of Items Reliability Source
Dependent Variables
Satisfaction
5
3
3
3
α = .90
α = .96
α = .98
α = .99
2
2
2
r = .72
r = .89
r = .75
Covariates
Age
Manipulation Check
Persuasiveness
Proactiveness
Product mix
86
Hui et al (2004)
DeCarlo (2005)
Maxham III and
Netemeyer (2002)
4.3.4 Results
4.3.4.1
Manipulation Checks and Item Reliability.
Manipulation check results showed that the experimental manipulation had
worked for all three different cues. The insurance agent’s reasoning was
considered more persuasive under conditions of high persuasiveness than under
low persuasiveness (MHighPers = 3.95, MLowPers = 2.31, F(1,260) = 78.79, p <
.001). Other than expected, a lower, but still significant effect was also revealed
for the other two independent variables, proactiveness (F(1,260) = 8.46, p <
.005) and product mix (F(1,260) = 5.84, p < .017). Similarly, the manipulation
was influenced by interaction effects of proactiveness/ persuasiveness (F(1,260)
= 3.92, p < .05) and product mix/persuasiveness (F(1,260) = 5.85, p < .017). No
such effects were measured for the interaction of proactiveness/product mix
(F(1,260) = .10, p > .75) and the interaction among all three variables (F(1,260)
= 2.09, p > .14).
A second manipulation check revealed that respondents in the high
proactiveness condition reported a considerably higher value for proactiveness
than those in the low proactiveness condition (MHighProact = 5.50, MLowProact =
1.91, F(1,260) = 421.52, p < .001). Again, some significant effects were also
measured for the independent variable product mix (F(1,260) = 4.67, p < .04)
and for the interactive effect of product mix and proactiveness (F(1,260) = 3.25,
p < .08). The data suggest no significant influence of persuasiveness (F(1,260) =
.50, p > .48), or of the interactions of persuasiveness/product mix (F(1,260) =
2.63, p > .10), persuasiveness/proactiveness (F(1,260) = 2.45, p > .11) and the
interaction of all three independent variables (F(1,260) = 1.21, p > .27).
Finally, respondents who had received a scenario treatment with balanced
product mix reported a greater value for balanced mix than those exposed to the
biased mix condition (MBalancedMix = 4.29, MBiasedMix = 2.12, F(1,260) = 182.28, p
< .001). As before, it was tested whether the other independent variables,
persuasiveness or proactiveness, or any interactions among the independent
87
variables had an effect on the manipulation check. The main effect of proactiveness (F(1,260) = 48.95, p < .001) was significant, as were the interaction effect
of proactiveness/product mix (F(1,260) = 6.10, p < .014) and the interaction of
all three independent variables (F(1,260) = 3.52, p < .07). No significance was
measured for the the main effect of persuasiveness (F(1,260) = .34, p > .56), or
the interaction effects of persuasiveness/proactiveness (F(1,260) = 2.27, p > .13)
and persuasiveness/product mix (F(1,260) = .11, p > .73). Overall, the manipulations in Experiment 2 must be deemed sufficiently effective, but not as precise
as desired.
Item reliability was assessed using the Cronbach’s alpha coefficient, which
yielded satisfactory values for the construct “customer-oriented attributions”
(0.90), for satisfaction (0.96), purchasing intention (0.98) and willingness to
provide WOM (0.99).
Due to the multiple dependent variables, a multivariate ANOVA was used to
test for the main and interactive effects of cues on customer perceptions and
intentions. Age was included as a covariate, since persuasion knowledge may
affect customers’ attributional thinking and the extent and sophistication of
customers’ persuasion knowledge has been argued to grow over their lifetime
(Friestad and Wright 1994). Correspondingly, younger customers have been
found to employ different and fewer response strategies to salesperson
persuasion than older customers (Kirmani and Campbell 2004).
4.3.4.2
Hypotheses Testing
Effect of persuasiveness on customer reactions. The results of my second
quantitative study confirmed that persuasiveness of reasoning had a positive
influence on customers’ satisfaction (Fsatis(1,260) = 57.20, p < .001), their
intention to purchase (Fpurchase(1,260) = 52.93, p < .001), and their willingness to
provide word-of-mouth (FWOM(1,260) = 53.95, p < .001). These values are also
displayed in Table 4-11.
88
Respondents were more satisfied when persuasiveness was high (Msatis = 3.48)
than when it was low (Msatis = 2.10). They were also more intent on purchasing
(Mpurchase = 3.18) and more willing to provide word-of-mouth (MWOM = 2.95)
under high persuasiveness than under low persuasiveness (Mpurchase = 1.88,
MWOM = 1.71). These results provide further support for H1 and are summarized
in Table 4-12.
reactions as dependent variable, main effects15
Dependent Variable
F(1, 260)
p
Persuasiveness of
Satisfaction
Reasoning (high/low) Intention to purchase
57.20
52.93
53.95
p < .001
p < .001
p < .001
Proactiveness
(high/low)
Satisfaction
25.03
15.64
23.63
p < .001
p < .001
p < .001
Product Mix
(balanced / biased)
Satisfaction
6.24
1.43
2.51
p < .014
p > .23
p > .11
Table 4-12: Mean values for customer reactions as dependent variables
in experiment 2, main effects 16
Low Persuasiveness
Dependent Variables
Satisfaction
Mean
2,10
1,39
1,88
1,36
Willingness to provide WOM 1,71
1,17
2,31
Manipulation Check
1,42
89
High Persuasiveness
Mean
3,48
1,53
3,18
1,49
2,95
1,56
3,95
1,59
Moderation of proactiveness. Similar to experiment 1, the data of the second
experiment highlight a significant interaction between persuasiveness of
reasoning and proactiveness for all three variables (Fsatis(1,260) = 6.63, p < .012;
Fpurchase(1,260) = 8.11, p < .006; FWOM(1,260) = 9.37, p < .003). The values are
also listed in Table 4-13.
reactions as dependent variables, interaction effects 17
Dependent Variable
F(1, 260)
p
Persuasiveness x
Proactiveness
Satisfaction
6.63
8.11
9.37
p < .012
p < .006
p < .003
Persuasiveness x
Product Mix
Satisfaction
3.91
8.85
12.86
p < .05
p < .004
p < .001
As table 4-14 shows, customer reactions were only improved through higher
proactiveness when persuasiveness was also high (high proactiveness:
Msatis = 4.08; Mpurchase = 3.73; MWOM = 3.55; low proactiveness: Msatis = 2.87;
Mpurchase = 2.64; MWOM = 2.34). These interaction effects are illustrated in Figure
4-11. Under conditions of low persuasiveness, this effect did not occur. Whether
proactiveness was high (Msatis = 2.32; Mpurchase = 1.98; MWOM = 1.86) or low
(Msatis = 1.90; Mpurchase = 1.78; MWOM = 1.57) made no significant difference.
90
experiment 2, interaction persuasiveness x proactiveness 18
Low Persuasiveness
High Persuasiveness
Low Pro- High Pro- Low Pro- High Proactiveness activeness activeness activeness
Dependent Variables
Mean
Mean
Mean
Mean
Satisfaction
1,90
1,30
1,78
1,34
1,57
1,08
1,96
2,32
1,46
1,98
1,38
1,86
1,26
5,29
2,87
1,35
2,64
1,40
2,34
1,40
1,86
4,08
1,46
3,73
1,39
3,55
1,47
5,69
1,46
1,55
1,47
1,21
Manipulation Check
customer reactions as dependent variables21
4,5
Satisfaction
4,0
3,5
3,0
Proactiveness
LOW
2,5
2,0
Proactiveness
HIGH
1,5
1,0
0,5
0,0
LOW
HIGH
91
Figure 4-11(cont.): Interaction of persuasiveness and proactiveness in
4,0
Purchasing
Intention
3,5
3,0
2,5
Proactiveness
LOW
2,0
1,5
Proactiveness
HIGH
1,0
0,5
0,0
LOW
HIGH
4,5
4,0
3,5
3,0
2,5
2,0
1,5
1,0
0,5
0,0
Willingness to
provide WOM
Proactiveness
LOW
Proactiveness
HIGH
LOW
HIGH
92
0.001 for all three customer reaction variables. For low persuasiveness, p-values
were > 0.09 (satisfaction), > 0.4 (purchasing intention) and > 0.16 (willingness
to provide WOM). The findings are fully consistent with H2 and reconfirm that
the effect of proactiveness on customer reactions is contingent upon the level of
persuasiveness.
Moderation of product mix. Hypothesis 3 postulated that the effect of product
mix on customer reactions was similarly contingent upon the level of persuasiveness. Indeed, as table 4-13 shows, experiment 2 yielded further evidence
that this cue’s influence on satisfaction, purchasing intention and willingness to
provide WOM was moderated by persuasiveness (Fsatis(1,260) = 3.91, p < .05;
Fpurchase(1,260) = 8.85, p < .004), FWOM(1,260) = 12.86, p < .001). Product mix
had a positive effect on customer reactions only when persuasiveness of
reasoning was high (balanced product mix: Msatis = 3.78; Mpurchase = 3.47; MWOM
= 3.28; biased product mix: Msatis = 3.11; Mpurchase = 2.84; MWOM = 2.54). Under
low persuasiveness, a similar interaction could not be observed (balanced
product mix: Msatis = 2.15; Mpurchase = 1.74; MWOM = 1.56; biased product mix:
Msatis = 2.06; Mpurchase = 2.03; MWOM = 1.87). These values can also be obtained
from Table 4-15.
0.05 for satisfaction, purchasing intention and willingness to provide WOM. For
low persuasiveness, p-values were > 0.7 (satisfaction), > 0.23 (purchasing
intention) and > 0.15 (willingness to provide WOM). Unlike experiment 1, the
second experiment delivered consistent results for all three customer reaction
variables, further corroborating Hypothesis 3.
93
experiment 2, interaction persuasiveness x product mix 19
Low Persuasiveness of
Reasoning
High Persuasiveness of
Reasoning
Biased
Prod. Mix
Balanced
Prod.Mix
Biased
Prod. Mix
Balanced
Prod.Mix
2,05
2,15
3,11
3,78
1,42
2,03
1,54
1,87
1,36
2,08
1,18
1,38
1,74
1,15
1,56
0,95
4,20
1,64
1,56
2,84
1,41
2,54
1,41
2,16
1,29
1,44
3,47
1,51
3,28
1,60
4,36
1,66
Dependent Variables
Satisfaction
Manipulation Check
customer reactions as dependent variables 23
4,0
Satisfaction
3,5
3,0
2,5
2,0
Product Mix
BIASED
1,5
1,0
Product Mix
BALANCED
0,5
0,0
LOW
HIGH
94
experiment 2; customer reactions as dependent variables 24
4,0
Purchasing
Intention
3,5
3,0
2,5
2,0
Product Mix
BIASED
1,5
1,0
Product Mix
BALANCED
0,5
0,0
LOW
HIGH
3,5
Willingness to
provide WOM
3,0
2,5
2,0
Product Mix
BIASED
1,5
1,0
Product Mix
BALANCED
0,5
0,0
LOW
HIGH
95
Up to this point, the analysis of Experiment 2 investigated a) the direct
relationships between cues and customer reactions and b) the interaction among
cues. As in the previous experiment, the research proceeded by focusing on the
mediating role of customer-oriented attributions. In order to validate the
hypotheses 4, 5 and 6, initial tests addressed the direct influence of the three
cues on customer-oriented attributions and, again, potential interaction effects.
Subsequent testing then incorporated a stepwise analysis of all three sets of
variables – cues, attributions and reactions.
Effect of persuasiveness on customer-oriented attributions. In preparation of
testing the hypotheses 4 to 6, the next analysis investigated whether the three
behavioral cues had similar effects on “customer-oriented attributions” as the
dependent variable. As in the previous experiment, the results (Table 4-16)
suggest a significant relationship between persuasiveness and customer-oriented
attributions (Fattr(1,260) = 59.76, p < .001). Respondents were more likely to
attribute favorable motives behind the advisor’s actions when persuasiveness
was high (MHI Persuas = 3.40) than when it was low (MLO Persuas = 2.29).
Table 4-16: Results of a univariate analysis in experiment 2; customeroriented attributions as dependent variable 20
F(1, 260)
p
Persuasiveness (High / Low)
Proactiveness (High / Low)
Product mix (Balanced / Biased)
59.76
33.23
1.10
p < .001
p < .001
p > .29
Persuasiveness x Proactiveness
Persuasiveness x Product Mix
9.66
3.73
p < .003
p < .06
Moderation of proactiveness and product mix. The initial results from an
univariate analysis of variances suggest a significant interaction between
persuasiveness of reasoning and proactiveness (Fattr(1,260) = 9.66, p < .003).
Under high persuasiveness, more favorable attributions were triggered when
96
proactiveness was high than when it was low (MHI Proact = 3.97; MLO Proact = 2.82).
When persuasiveness was low, the effect was considerably smaller, but still
significant (MHI
Proact
= 2.49; MLO
Proact
= 2.12). T-testing delivered values for
high persuasiveness t(136) = -6.10, p < .001) and low persuasiveness t(124) =
-1.90, p < .06).
Results also suggest a smaller, but still significant moderation of product mix
through persuasiveness (Fattr(1,260) = 3.73, p < .06), be it under conditions of
high persuasiveness (MBalanced
persuasiveness (MBalanced
Mix
Mix
= 3.55; MBiased
= 2.23; MBiased
Mix
Mix
= 3.22), or under low
= 2.36).
However, such an
influence is not supported by the planned contrasts that were performed
subsequently. Product mix had no significant influence on customer oriented
attributions, be it under conditions of high persuasiveness (t(136) = -1.55, p >
.12), or under low persuasiveness (124) = 0.62, p > .54). The mean values are
listed in Table 4-17, the interaction effects are illustrated in Figure 4-13.
variables in experiment 2 21
Dependent Variables
Low Persuasiveness
2,29
1,10
Low Persuasiveness
High Persuasiveness
3,40
1,23
High Persuasiveness
Low Pro- High Proactiveness activeness
2,12
2,49
1,06
1,12
Low Persuasiveness
Low Pro- High Proactiveness activeness
2,82
3,97
1,03
1,16
High Persuasiveness
Biased
Balanced
Prod. Mix Prod. Mix
2,36
2,23
Biased
Balanced
Prod. Mix Prod. Mix
3,22
3,55
1,15
1,31
1,06
97
1,15
Figure 4-13: Interaction of persuasiveness and proactiveness / product mix in
experiment 2; customer-oriented attributions as dependent variable 25
4,5
Proactiveness
4,0
3,5
3,0
Proactiveness
LOW
2,5
2,0
Proactiveness
HIGH
1,5
1,0
0,5
0,0
LOW
HIGH
4,0
Product Mix
3,5
3,0
Product Mix
BIASED
2,5
2,0
1,5
Product Mix
BALANCED
1,0
0,5
0,0
LOW
HIGH
Mediation of persuasiveness through customer-oriented attributions. After the
previous tests had confirmed significant relationships between the three cues on
one side and customer-oriented attributions on the other side, experiment 2
proceeded analogue to the first experiment. The potential role of attributions as
98
mediator was investigated through a series of regression analyses, as proposed
by Baron and Kenny (1986). The results, as illustrated in Figure 4-14, provide
further support for H4 which forwards the idea that customer-oriented
attributions mediate the influence of persuasiveness on customer reactions.
Firstly, persuasiveness (dummy variable: low = 0, high = 1) had an impact on
the dependent variables (ȕsatisf = .43, p < .001; ȕpurchase = .42, p < .001; ȕWOM =
.41, p <.001). Secondly, persuasiveness was also related to customer-oriented
attributions (ȕattr = .43, p < .001). Thirdly, customer-oriented attributions were a
significant predictor of the dependent variables (ȕsatisf = .74, p < .001; ȕpurchase =
.77, p < .001; ȕWOM = .73, p <.001). Lastly, when both persuasiveness and
customer-oriented attributions were included in the regression model, the
mediator remained a significant predictor (ȕsatisf = .69, p < .001; ȕpurchase = .73, p
< .001; ȕWOM = .68, p <.001), whereas the impact of the independent variable
decreased (ȕsatisf = .13, p < .005; Sobel: z = 6.76; ȕpurchase = .11, p < .02; Sobel: z
= 6.91; ȕWOM = .12, p <.02; Sobel: z = 6.70). Thus, H4 is confirmed.
attributions in experiment 2 26
0.43
Customeroriented
attributions
Persuasiveness
of reasoning
WOM: 0.68 (0.73)
Customer
reactions
Puchasing Int.: 0.11** (0.42)
WOM: 0.12** (0.41)
given outside the parentheses. p was significant at < .001 level, < .005 (*) or .02 (**)
99
Mediation of proactiveness through customer-oriented attributions. The analysis
of interaction effects among the three cues had consistently shown in
experiments 1 and 2 that proactiveness influenced customer reactions only under
conditions of high persuasiveness. In order to test whether the influence of
proactiveness on customer reactions was mediated, two separate analyses were
performed, splitting the sample into high (n=136) and low persuasiveness
(n=124) groups. The procedure was identical to experiment 1. Firstly,
proactiveness (dummy variable: low=0, high=1) had an impact on the dependent
variables when persuasiveness was high (ȕsatisf = .40, p < .001; ȕpurchase = .37, p <
.001; ȕWOM = .39, p <.001), but had no such effect under low persuasiveness
(ȕsatisf = .15, p > 0.09; ȕpurchase = .07, p > 0.4; ȕWOM = .13, p > 0.16). Hence, no
mediation was possible when persuasiveness was low. Proactiveness was also
related to customer-oriented attributions (low persuasiveness: ȕattr = 0.17, p
<.06; high persuasiveness: ȕattr = .47, p < .001). Thirdly, customer-oriented
attributions were a significant predictor of the dependent variables (low
persuasiveness: ȕsatisf = .72, p < .001; ȕpurchase = .78, p < .001; ȕWOM = .73, p
<.001; high persuasiveness: ȕsatisf = .66, p < .001; ȕpurchase = .68, p < .001; ȕWOM =
.64, p <.001). Lastly, when both proactiveness and customer-oriented
attributions were included in the regression model, the mediator remained a
significant predictor (Low persuasiveness: ȕsatisf = .72, p < .001; ȕpurchase = .79, p
< .001; ȕWOM = .73, p <.001; High persuasiveness: ȕsatisf = .61, p < .001; ȕpurchase =
.65, p < .001; ȕWOM = .58, p <.001), whereas the impact of the independent
variable was eliminated (low persuasiveness: ȕsatisf = .03, p > .60; Sobel:
z=1.867; ȕpurchase= -.06, p > .25; Sobel: z = 1.875; ȕWOM = .003, p > .95; Sobel: z
= 1.868; high persuasiveness: ȕsatisf = .11, p > .10; Sobel: z = 4.90; ȕpurchase = .06,
p > .39; Sobel: z = 5.08; ȕWOM = .12, p >.10; Sobel: z = 4.80. The relationships
are depicted in Figure 4-15. Consistent with H5, it can be concluded that under
high persuasiveness, the effect of proactiveness on customer reactions is fully
mediated by customer-oriented attributions.
100
Figure 4-15: Mediation of proactiveness through customer-oriented
attributions (under conditions of high persuasiveness) in experiment 2 27
0.47
Customeroriented
attributions
WOM: 0.58 (0.64)
Customer
reactions
Proactiveness
Satisfaction: n.s.* (0.40)
Puchasing Int.: n.s.** (0.37)
WOM: n.s.* (0.39)
given outside the parentheses. p was significant at < .001 level. And not significant at >.10
(*) or .375 (**)
Mediation of product mix through customer-oriented attributions. A final
analysis examined if customer-oriented attributions also mediated the impact of
product mix. Again, it was necessary to perform separate analyses for the two
persuasiveness conditions since the relationship between product mix and
customer reactions should be positive in the high persuasiveness and
insignificant in the low persuasiveness conditions. As expected, the impact of
product mix (dummy variable: low=0, high=1) on the dependent variables was
contingent on the level of persuasiveness (low persuasiveness: ȕsatisf =.04, p >
.70; ȕpurchase = -0.11, p > .2; ȕWOM= -0.13, p > .15; high persuasiveness: ȕsatisf
=.22, p < .01; ȕpurchase =.21, p < .015; ȕWOM =.24, p <.006). Hence, only under
high persuasiveness, a first critical prerequisite of any mediation was given.
However, the regression analysis yielded little evidence for any relation between
product mix and customer-oriented attributions (low persuasiveness: ȕattr = 0.06, p >.50; high persuasiveness: ȕattr =.13, p > .125). This is in line with the
planned contrasts performed earlier which found no significant relationship
101
between product mix and customer-oriented attributions. Customer-oriented
attributions were a significant predictor of the dependent variables (low
persuasiveness: ȕsatisf = .72, p < .001; ȕpurchase = .78, p < .001; ȕWOM = .73, p
<.001; high persuasiveness: ȕsatisf =.66, p < .001; ȕpurchase =.68, p < .001;
ȕWOM=.64, p <.001). Lastly, when both product mix and customer-oriented
attributions were included in the regression model, the mediator remained a
significant predictor (low persuasiveness: ȕsatisf =.73, p < .001; ȕpurchase =.78, p <
.001; ȕWOM =.72, p <.001; high persuasiveness: ȕsatisf =.64, p < .001; ȕpurchase =.67,
p < .001; ȕWOM =.62, p <.001), whereas the impact of the independent variable
was diminished (low persuasiveness: ȕsatisf = 0.08, p > .20; Sobel: z= -.62;
ȕpurchase = -.06, p > .25; Sobel: z=-.62; ȕWOM = -.09, p > .15; Sobel: z=-.62; high
persuasiveness: ȕsatisf =.14, p < .04; Sobel: z=1.53; ȕpurchase =.12, p <.06; Sobel:
z=1.53; ȕWOM =.16, p <.02; Sobel: z=1.53). The insignificance of the relationship
between product mix and customer-oriented attributions seems too marginal to
conclude that – with regard to hypothesis 6 – the results of experiment 2 are in
sharp contrast of those achieved in experiment 1. After all, all other relationships
in the mediation model depicted in Figure 4-16 show values that are in a similar
bandwidth to those found in the previous experiment (see also Figure 4-8). In
any case, in contrast to Study 1, the results of Study 2 do not support H6.
102
Figure 4-16: Mediation of product mix through customer-oriented
attributions (under conditions of high persuasiveness) in experiment 2 28
n.s.
(p>.125)
Customeroriented
attributions
WOM: 0.62 (0.64)
Customer
reactions
Product mix
Satisfaction: 0.14** (0.22)*
Puchasing Int.: 0.12** (0.21)*
WOM: 0.16** (0.24)*
given outside the parentheses. p was significant at < .001 level., < .015 (*) or .06 (**).
4.3.5 Discussion
Study 2 confirmed that in an open-architecture sales context, the persuasiveness
of reasoning, the proactiveness in offering third-party products and the mix of
recommended products are powerful cues that are perceived by clients and affect their reactions. Using a different medium for the experimental stimuli and a
different services category from Study 1, Study 2 provides consistent results
pertaining to the moderating role of persuasiveness on other behavioral cues.
With the exception of H6, it also supports the mediation of those cues’ effects
through customer-oriented attributions. Thus, it bolsters the generalizability and
robustness of the hypotheses forwarded in this dissertation. Two relevant points
of difference between the detailed results of the two experiments must be noted.
Firstly, unlike the first experiment, the second one reveals an interaction effect
between persuasiveness and product mix for all customer reactions, including
purchasing intention. Thus, it provides even broader evidence of the general
moderating influence that persuasiveness has on the two other cues. On the other
103
hand, no direct influence of product mix on customer-oriented attributions was
found. Consequently, the hypothesis that these attributions mediate the effect of
product mix on customer reactions is supported by the first, but not the second
study.
4.4 Influence of Customer Expertise
The research presented in this dissertation focuses on how customers evaluate
the motives behind a specific form of product offering and on the reactions they
subsequently display. As with almost any experimental research, there will be a
wide variety of factors which have not been addressed by the conceptual model
but may have a potential influence on these reactions. One of the more obvious
factors might be the product expertise that customers themselves bring to the
table. A customer’s own expertise with a product or service category has been
found to frequently influence attitudes and behaviors that occur in a sales context (e.g., Alba and Hutchinson 1987; de Bont and Schoormans 1995; Rao and
Monroe 1988); often in a moderating role (Bell, Auh, and Smalley 2005; Bell
and Eisingerich 2007; Sharma and Patterson 2000). In line with the definition of
Bell, Auh, Smalley (2005, p. 174), expertise is regarded here as "the extent of a
customer's prior product knowledge and ability to assess product performance."
Rao and Monroe’s (1988) research suggests that customers’ prior knowledge
about a product strongly affects their usage of specific cues when evaluating the
product. Sharma and Patterson (2000) argue that the relevance of trust in
building the advisor-client relationship decreases with growing expertise on the
client's side. This appears plausible: The more confident clients are in their own
knowledge, the less they will feel dependent on the advisor – and the advisor’s
trustworthiness. Moreover, the attributions that customers with varying levels of
product knowledge generate may be different ones. With regard to financial
investments, for instance, clients often struggle to evaluate the quality of finan-
104
cial advice they have received. One reason for this is that they lack the expertise
and experience to compare the potential outcomes of alternative financial
planning scenarios (Sharma and Patterson 1999). Clients with considerable
knowledge of investment matters may see things differently. For instance, they
tend to prioritize aspects of technical service quality in their judgment, as they
are capable of evaluating the actual core offering, e.g., the performance of
certain investment funds (Bell and Eisingerich 2007).
In order to shed some light on this relationship, additional analyses were performed on the data of experiments 1 and 2. The data sets of both experiments
were each separated into two equally sized groups, split at the median of the
variable “customer expertise”. Three items measuring customer expertise had
been taken from Sharma and Patterson (2000) who had employed them in
assessing clients’ product norm experience, but they have also been used to
measure client expertise (Bell and Eisingerich 2007). The items focus strongly
on the service elements provided through the advisor, while the present experiments revolve around the actual product offering made by the advisor. Therefore
the number of items was extended by one that addresses clients’ product familiarity. The resulting 4-item construct yielded alphas of .92/.94 (Experiment 1/2).
Table 4-18: Influence of customer expertise on customers’ attributional
thinking and reactions 22
Experiment 1
Experiment 2
F(1, 238)
p
F(1, 260)
p
0.51
0.69
1.37
2.92
> .47
> .41
> .24
> .08
0.04
0.22
0.45
0.27
> .84
> .63
> .50
> .61
Dependent Variables
Customer-oriented attributions
Satisfaction
105
As can be derived from the values in Tables 4-18 and 4-19, none of the experiments provided any evidence for a significant influence of customer expertise.
Table 4-19: Influence of customer expertise on customers’ attributional
thinking and reactions (mean values) 23
Experiment 1
Low
Expertise
Experiment 2
High
Expertise
Low
Expertise
High
Expertise
Dependent Variables
Customer-oriented attributions
Satisfaction
2.57
1.11
2.61
1.55
2.80
1.54
2.20
1.46
2.48
0.91
2.45
1.41
2.57
1.48
1.90
1.29
2.89
1.16
2.87
1.55
2.63
1.51
2.41
1.47
2.86
1.42
2.77
1.68
2.49
1.63
2.31
1.57
These results are not necessarily inconsistent with the research cited above.
Those earlier studies have argued, for instance, that customer expertise
moderates the influence of technical and functional service quality on
satisfaction (Bell et al. 2005; Bell and Eisingerich 2007). The data of the present
dissertation do not contradict such findings, but merely suggest that customers’
reaction to the specific salesperson cues proactiveness, product mix and
persuasiveness are not influenced by their expertise. Additional thoughts on this
matter are discussed in the final chapter of this dissertation.
106
5
DISCUSSION
This chapter will present a final discussion of the results and implications of this
dissertation. It is structured as follows: The first section will briefly reiterate the
reasoning, hypotheses and results for the two quantitative studies. The second
part will outline the theoretical and managerial contributions of this dissertation,
followed by a discussion of the limitations of the studies. The dissertation will
conclude by making a number of suggestions for future research.
5.1 Summary of Results
Selling competitor products through one's proprietary distribution channels
would seem a double-edged business strategy. Its obvious risks will need to be
outweighed by strong and positive overall customer reactions. Against this
background, it is surprisingly difficult to find any consistent and strong evidence
of the benefits and implications of an open-architecture offering, both in
academic and in managerial literature. Therefore, this dissertation has set out to
answer a number of research questions that were formulated in the introductory
chapter:
1. How do customers react towards an open-architecture offering?
Are there specific behavioral cues that salespeople provide during
the sales episode that will influence these customer reactions?
2. To what extent, if any, is such an influence subject to interaction
effects among the different cues?
3. Lastly, is the relationship between salesperson cues and customer
reactions mediated by attributions that customers generate in
order to explain the salesperson's behavior?
107
In order to evaluate whether the present research has provided satisfactory
answers to these questions, its main findings will be summarized in the
following paragraph. Building on existing consumer research literature, a
conceptual model was developed. Firstly, the model postulates that different
behaviors displayed by salespeople affect customer reactions towards an openarchitecture offering. It also forwards the idea that among these behavioral cues,
interaction effects may occur. Finally, the model claims that the relationship
between salesperson cues and customer reactions may not always be direct, but
that cues also influence the attributional thinking of customers, which in turn
affects their reactions.
A series of qualitative interviews helped to further explore the postulated
relationships and to identify specific factors of influence relevant in an openarchitecture context. As one result, the qualitative study yielded a variety of
different and distinctive attributions, of which the majority were customerrather than suspicion-oriented. The latter finding was in line with the assumption
that customers’ preference for variety and the counter-intuitiveness of an open
product architecture would support a favorable perception of the offering.
Furthermore, the interviews helped to identify three relevant behavioral cues
that customers seem to evaluate when making causal inferences about an open
product architecture: the proactiveness of the salesperson in offering competitor
products, the ‘balance’ of in-house vs. external solutions in the recommended
mix of products and, finally, the persuasiveness of the salesperson’s reasoning
behind each product. These cues do certainly not represent an exhaustive list of
all stimuli that are available to and evaluated by customers over the course of a
sales episode. However, they are more specific to the scenario of an openarchitecture offering than others (such as salesperson appearance, similarity or
messaging style), and for this reason, they were chosen to be further investigated
in this dissertation. Based on findings from the literature review and the
qualitative study, a number of hypotheses were forwarded and tested in form of
two experiments.
108
Experiment 1 demonstrated that in an open-architecture sales context, the
persuasiveness of reasoning, the proactiveness in offering third-party products
and the mix of recommended products are powerful cues that are perceived by
clients and affect their reactions. While persuasiveness of reasoning had a
significant direct effect on customer reactions, proactiveness and product mix
only had a positive effect on customer reactions when at the same time persuasiveness was high. In other words, if a salesperson disappoints in providing a
convincing explanation behind his or her choice, the proactive offering of thirdparty products or a balanced mix of in-house and third-party products will not
save the day. These findings support the hypotheses 1, 2 and 3. The results from
Study 1 also suggest that the influence of all three cues on customer reactions is
substantially and, in some cases, fully mediated by customer-oriented
attributions. For proactiveness and product mix, this effect is contingent on the
level of persuasiveness, which is in line with the interaction effects previously
highlighted. Consequently, hypotheses 4, 5 and 6 could be accepted. In order to
heighten the generalizability of results and demonstrate their robustness, the
second experiment investigated the same relationships in another setup. Using a
different medium for the experimental stimuli, an alternative services category
and a different sampling approach from Study 1, Study 2 provides consistent
results pertaining to the influence of the three behavioral cues, the moderating
role of reasoning persuasiveness on the other two cues and on the mediation of
these effects through customer-oriented attributions. With the exception of H6,
which was not supported, Study 2 confirms the validity of the first study’s
findings. A final analysis was performed in order to investigate a potential
influence of customers’ own expertise (with investment funds / insurance) on
customer attributions and reactions. The fact that no significant effects were
found would seem to contradict existing research that claims a substantial
relationship between customer expertise and product evaluations. However, it
seems more likely that only for customers’ interpretation of the specific cues
tested in the present research, customer expertise had little relevance.
109
5.2 Theoretical contribution
Up to this point, the specific outcomes of the dissertation research have been
discussed and summarized. The following chapter will therefore outline the
more general implications that the present studies contribute to different lines of
marketing research.
5.2.1 Contribution to Literature on Open Product Architectures
First of all, this dissertation aims to shed light onto the behavioral and cognitive
consequences triggered by a sales approach that has been hardly researched
before. The fact that firms, like banks or own-label retailers, open their
proprietary distribution channel(s) to competitor products, has received
considerable attention in the consumer press and industry publications.
However, the effects of such an open-architecture offering have, to the author's
knowledge, not been subject of any consumer behavioral research before. A
number of conclusions can be drawn from fields of study that bear a certain
resemblance or relevance to the topic – such as counter-intuitive sales
messaging, disconfirmed expectancies and attribution theory in general. These
can help to understand isolated aspects of how an open architecture may affect
customer perceptions and actions, but they do not allow to gain a comprehensive
and integrative understanding of the different effects at work. The present
dissertation attempts to close this gap by proposing a specific conceptual model
that addresses both cognitive and behavioral reactions customers are likely to
display when faced with an open-architecture offering. Specifically, it outlines
how salesperson behavior and customers' attributional thinking interact in
influencing these reactions. The model ties back to a broad range of earlier
findings from marketing and services research. In addition, it has been refined
based on a qualitative study that explored actual bank customers' views on
financial advice and their interpretations of an open product architecture. In
order to reflect the likely complexity of causal relationships, care was taken to
110
account for both mediation and moderation effects. The hypothesized
relationships articulated by the conceptual model have been empirically tested in
form of two quantitative experimental studies. In spite of certain limitations that
will be discussed later in this chapter, the results suggest a satisfactory validity
of the conceptual model.
5.2.2 Contribution to Literature on Cue Influence in Selling
This dissertation underlines the influence that salesperson behavior observed
during a sales episode has on customers' reactions, adding to a considerable
body of existing evidence (e.g. Bell et al. 2005; Bell and Eisingerich 2007;
Doney and Cannon 1997; Hawes et al. 1989; Kennedy et al. 2001; Pornpitakpan
2004; Sirdeshmukh et al. 2002; Sparks and Areni 2002; Sujan et al. 1986; Swan
et al. 1999; Weitz 1981; Weitz et al. 1986; Wood et al. 2008a; Wood et al.
2008b). Specifically, the present work extends our understanding of the effects
of counter-intuitive selling approaches. Past research on the effects of counterintuitive selling has often focused on communication messages that seem
incongruent with the communicator's expected motivations and beliefs. Typical
experiments have featured, for instance, an individual that advocates an
unexpected opinion (Kohn and Snook 1976; McPeek and Edwards 1975) or a
salesperson or advertisement that points out not only the strengths, but also
some weaknesses of a promoted product (Etgar and Goodwin 1982; Golden and
Alpert 1987; Pechmann 1992). Key findings in this field have been that
consumers often perceive seemingly counter-intuitive behaviors and messages
from salespersons as signs of benevolence (Sirdeshmukh et al. 2002) or
'authentic' product information (Hunt et al. 1982), and that the advocacy of an
unexpected position can improve a communicator’s credibility (Crowley and
Hoyer 1994; Eisend 2007; Etgar and Goodwin 1982; Golden and Alpert 1987;
Koeske and Crano 1968; Kohn and Snook 1976; Pechmann 1992). The research
presented in this dissertation supports these earlier findings. More importantly,
111
though, it extends the existing body of evidence by focusing on counter-intuitive
stimuli other than the sales message itself.
For one, the qualitative pre-study provides a first indication that an openarchitecture offering in itself represents a case of counter-intuitive selling – the
majority of interviewees expressed their astonishment at such a sales approach
and their likelihood to ‘wonder why’. More importantly, the experimental
findings clearly show that in the context of an open architecture, a customer's
observation of two specific counter-intuitive cues can lead to higher satisfaction,
purchasing intention and willingness to provide word-of-mouth. The first one,
proactiveness, is given when a salesperson offers competitor products at his or
her own initiative, rather than only following the customer's explicit request.
The second one, a balanced product mix, is provided when the salesperson's
recommendation considers in-house and competitor products in equal parts, i.e.,
without demonstrating an obvious bias for one or the other side. However, the
occurrence of both effects is contingent upon the level of persuasiveness that the
advisor's reasoning offers.
The qualitative study also hints at the negative effects that too strong counterintuitive signals may have. Asked what they would think if the advisor presented
them only with competitor products, the interviewees responded with
predominately negative inferences. Several respondents saw such a behavior as
an indication of the in-house products' low quality. A similar risk has been
pointed out for two-sided advertisements that highlight a product's weak points
too prominently: Crowley and Hoyer (1994) argue that at some point, the
credibility gained by exposing a product's weaknesses may be more than
outweighed by detrimental effects on a customer's actual intention to buy the
product. One respondent in the qualitative study saw a recommendation of only
external products as the advisor's all-too-obvious attempt to appear objective.
This suggests that such an approach could be perceived as a deliberate
persuasion tactic and thus trigger unfavorable responses (Friestad and Wright
1994; Tormala and Petty 2004). The latter negative findings help to illustrate the
112
richness of potential reactions to an open-architecture offering, but they were not
tested as part of the experiments and should therefore be treated with caution.
Lastly, the research presented in this dissertation underlines the importance of
investigating cue effects not in an isolated, but integrative manner. It has been
criticized that a considerable amount of cue research examines isolated effects,
while in reality, consumers will mostly form their opinion based on a variety of
cues. On their specific subject of product quality, Purohit and Srivstava (2001, p.
123) argue: “although consumers are exposed to multiple cues simultaneously in
the marketplace, there is relatively little understanding of how these cues are
combined or integrated in assessments.” Thus, by illustrating the moderation
effects among proactiveness, product mix and persuasiveness, a more integrated
view on cue effects is supported.
5.2.3 Contribution to Literature on Customers' Attributional Thinking
In addition to investigating the influence of cues, the present research also sheds
further light on the specific role of attributional thinking in the attitudeformation process that leads to customer reactions. Firstly, the qualitative
prestudy hints at the function of expectancy disconfirmation as a trigger of
customers' causal searches. Secondly, by listing a typology of various customeroriented and suspicion-oriented attributions, it provides a clear and tangible
impression of attributions as they are actually generated by customers. Thirdly,
the quantitative results of two experiments show a highly significant relationship
between customer-oriented attributions and customer reactions, confirming the
substantial influence that attributional thinking has been claimed to have. But
most importantly, the experimental results contribute to a better understanding
of the overall relationship between cues, attributions and customer reactions.
Earlier research has explored attributional thinking and also its link to the
disconfirmation of consumers’ expectancies (e.g. Hastie 1984; Hunt et al. 1982;
McPeek and Edwards 1975; Pyszczynski and Greenberg 1981; Smith and Hunt
1987; Sujan et al. 1986). However, only few studies have jointly investigated the
113
interaction of cues and attributions in instigating customer reactions. This thesis
extends previous findings of DeCarlo (2005) and supports the causal
relationships the author has established. It provides detailed evidence of the
mediating role of attributions in the context of an open-architecture offering. In
doing so, the dissertation highlights both potential antecedents and consequences
of attributional thinking and thereby helps us to gain a more holistic
understanding of this cognitive process.
In addition, some individual points are worth noting. Firstly, the experiments
have shown both direct and indirect effects on customer reactions for all three
cues – even though these were provided in different phases of the sales
encounter. This would support the idea that customers use a variety of cues
(Burnkrant 1975) and form their opinion along the whole sales episode, rather
than making up their mind after observing the first cue. Moreover, the last cue in
the experiment – persuasiveness of reasoning – turned out to be the one that
moderated the influence of the other previous two, proactiveness and product
mix. They had no positive impact on customer reactions unless persuasiveness
of reasoning was high. Put differently, it would seem that the positive motives
associated with high proactiveness or a balanced product mix were discounted
as long as the advisor's low persuasiveness suggested a lack of expertise or
interest in the customer's benefit.
5.3 Managerial Implications
The results of this dissertation have a number of implications for managers of
companies that either have an open-architecture offering in place or intend to
introduce such a sales model to their customers. The following sections will
review these implications, starting with those that are most directly linked to the
original research questions and the corresponding findings. Specifically,
consequences for the management and education of a sales force are outlined.
Subsequently,
the
wider
implications
114
for
areas
such
as
marketing
communications and brand management are pointed out, leading from particular
to more general considerations.
5.3.1 Delivering an Open-Architecture Offering to Clients
In order to improve their sales effectiveness, companies that employ an openarchitecture sales model will want to instigate favorable customer reactions
towards such an offering, and they will want to avoid negative ones. With this
goal in mind, the present dissertation has explored whether there are specific
cues that influence customer perceptions and intentions in a personal-selling,
open-architecture context. Moreover, different qualitative and quantitative
methods have been used to understand the cognitive processes that lead from the
observation of a cue to the attribution of causes and onwards to a customer
reaction. The research findings suggest that, indeed, there are specific behaviors
that a salesperson can demonstrate in order to instigate positive customer
reactions. A major part of these effects is indirect, i.e., the cues trigger favorable
customer-oriented attributions, which in turn lead to improved satisfaction,
purchasing intention and willingness to recommend. There is clear evidence
that both a proactive offering of third-party products and a balanced mix of inhouse versus external products are appreciated by customers. However, if the
sales agent fails to explain in a convincing and transparent manner why
individual products – and especially the third-party ones – have been selected,
neither proactiveness nor a balanced product mix can make up for this blunder.
These conclusions have several critical implications for the management of an
open or "guided" architecture offering, and possibly, for any sales approach in
which own-manufactured products are sold next to third-party ones.
115
5.3.1.1
Sales Force Competence
Firstly, companies that expect their sales force to successfully sell not only inhouse, but also third-party products need to make sure that their sales agents
have the necessary competence to provide a persuasive reasoning for these
products. Salesperson competence is a major driver of customer trust, and
according to Swan, Bowers and Richardson (1999) is perceived as displayed
through skills, ability and expertise. The importance of salesperson expertise in
building customer trust and improving the buyer-seller relationship has been
highlighted by numerous studies (e.g. Crosby et al. 1990; Doney and Cannon
1997; Wood et al. 2008a; Wood et al. 2008b). A company may be accustomed
to building sales force expertise through product trainings and seminars – but
not necessarily on the topic of third-party products. Hence, one challenge will be
to offer a comprehensive education program that ensures sales agents to be just
as knowledgeable about third-party products as they are about in-house
offerings. This should prove difficult, if not impossible, since the amount of
external products is – in categories such as investment products – virtually
infinite. Most likely, firms will therefore have to chose carefully which thirdparty products are offered proactively and then provide a sufficient training for
those.
Similar considerations apply for the competence dimensions of skill and ability.
The findings of this dissertation suggest that the persuasiveness of a
salesperson's reasoning may be one of the critical cues that determine the
ultimate success or failure of an open-architecture sales encounter. This is in line
with a considerable amount of research that has highlighted the importance of
selling skills such as presentation techniques, clear language and effective
messaging (e.g., Anselmi and Zemanek Jr 1997; Dion and Notarantonio 1992;
Plouffe et al. 2009; Rentz et al. 2002). In order to fully capitalize on the positive
expectancy disconfirmation that an open-architecture offering may trigger,
companies must make sure that their sales agents have the right skills to explain
the benefits of this sales model. From a customer's point of view, the need for a
116
clear overview and a convincing reasoning and recommendation is likely to
grow with the amount of options available. Moreover, salespersons must be able
to convincingly explain why in some cases a third-party product is better suited
than an in-house one, and do so without tarnishing their own company's brand.
In conclusion, strong selling skills will be even more urgently needed in an
open-architecture context than in other sales encounters.
The third dimension of competence as forwarded by Swan, Bowers and
Richardson (1999) is ability. Ability is demonstrated by executional excellence,
as opposed to expertise, which is based on superior knowledge (Sirdeshmukh et
al. 2002). If salespeople are expected to genuinely deliver an open-architecture
offering to their clients – i.e., demonstrate the ability – then this has a substantial
impact on the processes and infrastructure at their command. For instance, firms
have to provide their salespeople with equally concise and accessible
information on the available third-party products as on their own, in-house
products. Similarly, the actual process of transferring ownership of the product
to the client should not differ in convenience, whether it is an in-house or a
third-party product: If the sale is straightforward in one case, but in the other, the
salesperson and customer have to go through complex paperwork full of legal
disclaimers, this will put competitor products at a clear disadvantage. In both
cases – availability of information and ease of transaction - there is an obvious
risk that advisors and sales agents will rather stick to in-house solutions because
these are what they know best or sell with greater ease.
5.3.1.2
Sales Force Incentivization
Where a sound and convincing reasoning for the recommended products is
given, the two other behaviors that were investigated can further improve
customer reactions. If a company’s customers know that an open-architecture
product range is in place – e.g., because the firm has advertised it – then the
company’s sales agents should offer it on their own initiative. If they do not,
117
customers may feel that potentially better options are withheld from them.
Similarly, companies that promote an open-architecture offering should make
sure that their customers experience this extended choice as genuine: If sales
agents manage to present a balanced mix of in-house and third-party products,
customers tend to show positive reactions. If, on the other hand, competitor
products are only added to the recommendation as an afterthought, then
customers are likely to discard the proposal as a fraud. Both cues bear
fundamental implications for the ways in which companies currently set their
sales targets and incentivize their sales force. If companies want their sales force
to proactively and genuinely offer competitor products, they have to make sure
that the sales targets and reward systems support this objective. In the banking
industry, for instance, this may not always be the case, as Chapter 1.2 of this
dissertation suggests. As long as banks indirectly incentivize their advisors to
sell in-house products (Rasch 2003) and allocate their clients' money to ownlabel funds because of higher profit margins (Ross 2010), it seems likely that
many advisors will not treat external products equally to their own. Open
architecture remains a concept that should be administered in careful doses,
though. Overly eager salespeople or advisors that strongly push the open
architecture and keep stressing the objectiveness of their advice may be
perceived as ‘trying too hard’. Such deliberate provision of a certain cue in order
to activate a consumer's use of specific (and favorable) decision heuristics could
provoke unfavorable counter-reactions from clients (Friestad and Wright 1994;
Tormala and Petty 2004).
5.3.2 Promoting an Open-Architecture Prior to the Sales Encounter
5.3.2.1
Advertising Communication
In line with the main research focus of this dissertation, the previous section has
summarized different salesperson behaviors that are helpful (proactiveness,
product mix) or even critical (persuasiveness of reasoning) in the successful
118
delivery an open-architecture offering to clients. It has been argued that in order
to foster such behaviors, companies must take great care to develop the
knowledge and selling skills of their sales force. They also have to make sure
their sales personnel have the ability to include third-party products in their
recommendation and to execute a sale of such products. Finally, it would seem
obvious that a firm’s sales incentives should not discourage its sales agents from
offering external products, if its open-architecture offering is expected to be
successful.
Beyond the effect of specific salesperson cues, the conceptual model has
forwarded the idea that an open-architecture offering per se is likely to be
perceived favorably by customers. This argument was based on previous
research into the appeal of greater assortment variety and, secondly, into the
effects of a positive disconfirmation of expectations. Indeed, results from the
qualitative pre-study suggest that many customers would be positively surprised
by being offered third-party products. Assuming that such a positive effect
exists, it seems perfectly understandable if companies try to actively promote
their open architecture: Chapter 1.2 has provided a number of examples for how
German banks advertised their open architecture at the time they first introduced
this offering to their home market. If an open architecture disconfirms many
customers’ long-held expectations and instigates favorable attributions towards
the company and salesperson, it appears logical to advance this effect from the
time of the sales encounter to a much earlier point in the acquisition process.
Otherwise, a valuable selling proposition would be wasted: As long as the
availability of third-party products is ‘revealed’ only during the sales episode, it
can merely improve the satisfaction of existing clients or of those prospects that
have proceeded quite far towards becoming a client; the appeal of such an
offering is not utilized in attracting new prospects.
Those companies that decide to leverage the positive appeal of an open
architecture in their communication and actively promote it – for instance,
through advertising – will create a specific expectation on the side of their
119
potential and existing clients. This expectation must be fulfilled, i.e., customers
need to experience the open product architecture as an authentic offering. If not,
the effect may be detrimental: customers who feel they have been deceived by
either misleading or false advertising are likely to distrust the sender and
develop a negative attitude towards advertising in general (Darke and Ritchie
2007). And, while many products and services in areas like the financial
industry have credence qualities, firms should not fall under the illusion that
when the moment of truth comes, clients are not able to tell whether the promise
of an open architecture has been met. Instead, the present research suggests that
many customers do form an opinion on whether they’ve been offered third-party
products and whether the offer was genuine, and they will draw corresponding
conclusions. Hence, before making it a subject of their marketing communications, companies that dispose of an open product architecture need to make
sure they also deliver it to their customers in practice.
Where a firm does decide to actively promote its open architecture to customers,
the advertised reasons and benefits behind such an offering should be carefully
considered. This dissertation has argued that customers will wonder about a
firm’s motives if it offers them competitor products, and they will search for
explanations. In line with this, the persuasiveness of reasoning provided by a
salesperson turned out to be a powerful cue in eliciting positive customer
reactions. It seems therefore plausible that just like the salesperson is expected
to reason why a certain product (and especially a third-party one) has been
recommended, any advertising must explain in a convincing way why a firm
would offer an open-architecture in the first place. This reasoning may prove to
be a rather delicate task: if, for instance, the promise of “best-in-class” products
is emphasized too strongly, this may suggest to some customers that the
company’s own in-house products are of inferior quality. Such brand-related
considerations will be addressed in the final sections of this chapter.
120
5.3.2.2
Brand Positioning
The above advertising example highlights that companies fundamentally change
their value proposition when they begin to sell competitor products through their
proprietary distribution channels. The introduction of an open architecture
transports a message to customers that needs to be carefully aligned with the
company's brand positioning. What is the primary role that the firm intends to be
associated with: that of a manufacturer or that of a retailer? In the qualitative
pre-study, several interviewees attributed an open architecture to a firm's
motivation to extend the depth and bandwidth of its product offering. A
common perception was that very few firms can credibly claim that their own
products are the best in every single product category of their portfolio. Hence,
it is perfectly understandable to clients that a company would source certain
products – e.g., those that require a very specific manufacturing expertise – from
external partners. In other words, customers are prepared to accept a company’s
hybrid function of both manufacturer and retailer, as long as the two are not
contradicting each other.
The emphasis that companies put on each of the functions may differ
considerably, as they find themselves faced with a general and strategic decision
– whether to seek a positioning as either a solution provider / retailer with
certain manufacturing capabilities, or as a product manufacturer with its own
distribution channel. In the first case, a firm should not promote its own product
brand as the best one on every possible dimension, as this would obviously
contradict the idea behind its "open" sales model. The firm may even consider to
brand its own products in a way that their connection to the corporate brand is
less obvious, thereby suggesting that the in-house offering is treated like any
other external provider. Examples would be DWS Investments, the mutual funds
label of Deutsche Bank, or Union Investment, the funds management branch of
the German ‘Volksbanken Raiffeisenbanken’. Similarly, firms that want to
strengthen the relationship between the corporate brand and the product brand
may opt for an obvious reference in the brand name. After its own fund business
121
for private investors, branded “Adig”, had lost ground after the introduction of
an open product architecture, Commerzbank renamed the business to
“Cominvest”, aiming to leverage the parent company’s brand equity
(Anonymous 2006a, b).
5.3.2.3
Impact of Third-Party Brands
Finally, it may not be enough for managers to consider how the overall promise
of an open architecture changes customer perceptions of their firm and its brand.
There are also several reasons why the association with a specific third-party
brand that is distributed by the company could affect the firm’s own brand. For
one, the results of this dissertation suggest that customers react to an openarchitecture offering with a rich set of attributions. It has also been shown that in
the development of such causal inferences, customers evaluate different cues. In
the present research, these cues were provided in form of certain types of
behavior displayed by the salesperson. It seems highly plausible that the brand
image of one or several third-party brands offered to a customer would serve as
another input to this evaluation process. After all, there is abundant evidence for
the fundamental role that brands take in consumer decision processes (cf. Keller
and Lehmann 2006). Further support for such an influence of third-party brands
can be gained from research into the relationship between retailer brands and
assortment brands. Jacoby and Mazursky (1984) have demonstrated that
customers who associate a retailer brand with a manufacturer brand tend to
average their perception of both brands - the weaker one’s image is improved,
while the stronger brand suffers. Moreover, it has been argued that individual
‘anchor’ brands on the one hand and the number of well-known brands in a
firm’s assortment have separate and significant effects on the retailer brand
(Porter and Claycomb 1997). A direct relationship is also underlined by Mulhern
(1997) who concludes from a research review that the quality and reputation of
the brands listed by a retailer affect the retailer brand directly, and not
necessarily only by improving the firm’s perceived assortment quality. Such
122
brand effects have not been researched in this dissertation. However, the present
findings have demonstrated significant relationships between cues, customeroriented attributions and customer reactions. If, for good reasons, it is assumed
that the brands of external products represent another relevant cue, then similar
effects must be considered. Consequently, managers should carefully evaluate
not only whether they open their proprietary distribution channels to other
parties, but also to which ones exactly.
In summary, managers should be aware of the manifold implications of an openarchitecture sales model, some of which have been highlighted by the findings
of this dissertation. Even though anecdotal evidence from several Swiss and
German banks suggests a different thinking, the opening of a proprietary
distribution channel to third parties is not like any other “me-too” extension of a
firm’s product assortment. In order for this particular sales strategy to be
successful, managers have to rethink their approach to providing the right
competence and incentives to their sales force. Moreover, they should consider
early on how the open product architecture alters their company’s value
proposition and how this must be reflected in marketing communications and
branding.
5.4
Limitations
The present research bears some of the typical limitations imposed by
experimental studies. Firstly, laboratory experiments allow the researcher to
quite effectively control a defined set of independent variables and therefore
generate results of considerable internal validity (Bateson and Hui 1992; Bitner
1990). However, by excluding an unknown variety of other potential influences,
the scenarios presented to provide certain stimuli will rarely manage to fully
recreate an authentic environment, i.e., the ‘real world’. The present work, for
instance, has focused on the effects of three distinct behavioral cues provided
123
during the sales episode. Most customers, though, do not enter a sales encounter
completely unprejudiced. They rather hold long-established, robust views - for
instance, about the motives that drive a salesperson (Campbell and Kirmani
2000; Friestad and Wright 1994). The data provide some support for this notion:
The level of ‘general mistrust’ towards banks that the test subjects were asked to
indicate during Study 1 turned out to have a significant negative relationship
with customer-oriented attributions and customer reactions. This allows the
assumption that a broader exploration of both customer pre-dispositions and
cues provided during the sales episode may yield further insights and reveal
their relative importance in customers’ evaluation of an open-architecture
offering.
A second factor that may limit the external validity of the research results lies in
the medium used to present respondents with the different cues. Experiment 1
employed written scenarios to depict the customer-salesperson encounter. Such
text stimuli offer the researcher effective means to control specific cues, but
have been criticized of lacking realism (Bateson and Hui 1992). Therefore,
experiment 2 employed video vignettes of the sales scenario, a simulation that
has been argued to provide greater realism (Grandey et al. 2005) and lead to
results of satisfactory external and internal validity (Baker et al. 2002; Bateson
and Hui 1992; Grandey et al. 2005). Still, both experiments bear the inherent
limitations of laboratory research and a confirmation of the experimental
findings in a more externally valid context would clearly be desirable.
Beyond the limitations common to many research contributions, it should be
noted that this dissertation has refrained from making a number of distinctions
that may allow an even more detailed understanding of certain relationships. The
first one concerns the differentiation of sales agent and company. Customers
who are faced with an open-architecture sales scenario may attribute certain
salesperson behaviors either to the salesperson him- or herself or to the company
that this person is working for. Consequently, attributions of responsibility (see
Chapter 2.2.1) for a certain action may differ. The qualitative interviews
124
produced a few statements along this line, with respondents expressing their
belief that bank advisors would prefer to sell in-house products because a) this
earned them a higher commission, or because b) they were forced to do so by
their company’s demanding sales targets. However, the overall evidence found
for such a distinction by the customer was not compelling enough to be reflected
in the conceptual model. The majority of responses suggest that the interviewees
did not strongly distinguish between the advisor’s and the bank’s specific role in
offering third-party products. On the one hand, this is supported by findings that
customers often tend to ‘generalize’ a salesperson’s behavior, perceiving the
sales agent and the firm or brand as one and the same (Crosby et al. 1990;
Wentzel, Tomczak, and Herrmann 2008). On the other hand, Wentzel (2009)
argues that a customer will be more likely to distinguish between a company and
its salesperson the more he or she depends on this person’s assistance. It seems
plausible that the number of product choices available represents one reason
why such a perceived ‘outcome-dependency’ would grow (another might be, for
instance, the perceived risk of a certain type of investment product). In other
words, if an open product architecture offers a vast assortment that only an
expert can ‘navigate’, many customers will feel reliant on the sales agent – and,
consequently, form a more individual impression of this person. Further research
could help to clarify these effects.
Finally, this dissertation does often use the terms ‘salesperson’ / ‘sales agent’
and ‘(client) advisor’ synonymously. The reason for this is that the present
research is focused on the very concrete example of an open product architecture
as it has been implemented by many banks. Hence, the frequent usage of the
term ‘advisor’. Doubtlessly, many financial institutions would object such a
synonymous use, pointing out that an advisor’s role – in contrast to a
salesperson’s – is not primarily to sell, but to advise clients on financial matters,
and ideally do so in a client-oriented, neutral way. However, both the
controversial discussion outlined in chapter 1.2 and results of the qualitative
study suggest that many bank clients do not perceive their advisor as a
125
completely neutral source of information, but are well aware of the sales
motives (and, therefore, conflicts of interest) that are rooted in the close
relationship of an advisor and his or her bank (Bolton et al. 2007; Rasch 2003).
Some of the positive ‘surprise’ created by an open-architecture offering would
seem to have its root in those very expectations. Many statements gathered in
the qualitative pre-study suggest that it was the selling motivation of advisors
that many interviewees saw disconfirmed by an open architecture. Concluding,
it seems fair to assume that many of the presented findings should apply to
encounters with ‘salespeople’ as much as with ‘client advisors’.
5.5 Future Research
This dissertation attempts to improve our understanding of how customers react
to an open-architecture product offering and what role salesperson behavior and
attributional thinking play in this context. To this point, specific research on
open-architecture selling is scarce and the author is not aware of any earlier
contributions that apply a consumer-behavior angle . And while there are other
sales approaches that bear a certain resemblance – such as competitor
collaboration or two-sided advertising – further investigations into different
aspects of open-architecture selling seem promising.
5.5.1 High versus low Customer Expertise
Chapter 4.4 has argued that the financial literacy of customers themselves could
be another factor that influences reactions to an open architecture. Expert
customers may show much greater appreciation for such an extended offering
since they are confident to understand the differences among products and
consequently can choose the right one. Less knowledgeable customers, on the
other hand, may be confused by an open architecture and thus feel an even
126
greater dependency on the advisor's counsel, leading them to disapprove of such
an extended, but potentially intransparent offering. The fact that the results of
the present dissertation did not provide any evidence for such a relationship does
not disprove its existence. On this issue, several points need to be considered.
First of all, the experimental results only indicate that customers’ own expertise
has no significant effect on their evaluation of the three specific salesperson
behaviors that have been analyzed here. The data do not suggest that customer
expertise has also little influence on other aspects of how customers perceive
and react towards an open-architecture offering. It is important to remember that
the entire research design – such as the experimental approach and the provided
stimuli – aimed to investigate specific customer attributions and reactions. It was
not intended to measure the influence of customer expertise and may, for
instance, have presented cues for whose interpretation the level of customer
expertise had little relevance. Other cues, such as an advisor’s attempt to
‘educate’ the client on investment basics may have led to more different
reactions, depending on the customer’s pre-existing financial knowledge.
It should also be noted that many of the previously cited studies on customer
expertise have employed a different research methodology. Several empirical
studies that have found a significant impact of customer expertise (Bell et al.
2005; Bell and Eisingerich 2007; Sharma and Patterson 2000) employed largesample surveys investigating actual customer experiences and behaviors over a
long time period. It seems plausible that the evaluations of the technical and
functional aspects of a product or service that test subjects have actually
experienced would depend more strongly on their individual expertise than
would their evaluation of a scenario observed during a laboratory experiment.
Hence, a more appropriate approach for testing the influence of customer
expertise may feature a large-scale survey among customers who have actually
been confronted with an open-architecture offering. While such a survey would
hardly allow to isolate individual cues – as has been done in this dissertation –, it
could yield more general insights as to whether and how customers’ expertise
affects their evaluation of an open product architecture.
127
5.5.2 Manufacturer Role versus Retailer Role
The research presented in this dissertation has underlined the importance of cues
in customers’ evaluation of a sales or service encounter. Beyond the three
behavioral cues that have been investigated in the experimental studies, there is
obviously a variety of other cues that may directly or indirectly influence a
customer’s perception of and reaction towards an open-architecture offering. In
the context of an open-architecture sales model, one such cue could bear a
special relevance, namely what could be called the firm’s main role in the
distribution chain. How do customers perceive a bank, a car dealership or an
airline’s travel agency? Are they primarily the distribution outlet of a specific
manufacturer brand – or are they a retailer of products from a great variety of
sources, their own brand only one among many? This question seems to carry
significant importance for at least two reasons: Firstly, customers usually
perceive the reputation of a manufacturing brand and of a retailer as two
separate and relevant cues. Secondly, it seems likely that those customers who
perceive a firm as a retailer rather than a manufacturer will also be less surprised
by its open architecture: the counter-intuitiveness of the offering diminishes, and
with it, possibly, the urge for attributional thinking.
Previous research has demonstrated that consumers’ evaluation of a product’s
quality is significantly influenced by the reputation of both a product’s
manufacturer and the retailer through which the product is purchased (Purohit
and Srivastava 2001). Among the two, manufacturer reputation may be the most
important cue in such an assessment (Dodds, Monroe, and Grewal 1991; Grewal
et al. 1998; Rao and Monroe 1988). However, as retailers provide the interface
between customers and manufacturers, their image also bears a weight, and
consumers consequently take both cues into consideration (Purohit and
Srivastava 2001). In their investigation of a) the influence and b) the interaction
of manufacturer and retailer reputation on quality judgments, some of the above
studies have indeed distinguished between a manufacturer that sells through a
retailer (Dodds et al. 1991) and one that sells its products through its own,
128
proprietary distribution channels (Boulding and Kirmani 1993; Purohit and
Srivastava 2001). Typical examples for the first case would be consumer goods
such as Mars chocolate bars or Gillette razors, while the IKEA furniture store is
an example of the second case. Among other results, it was found that consumer
perceptions of product quality were higher when the manufacturer sold its
product through a reputed retailer relative to selling directly (Purohit and
Srivastava 2001). However, such a distinction does not cover the unique
characteristics of an open-architecture product offering. In the case of an open
architecture, the manufacturer is more than just its own distributor: Many banks,
for example, are well established manufacturing brands (of services as much as
of financial products), and at the same time, they take over a role of a fullyfledged retailer that also sells third-party and even competitor products. How do
customers perceive such a company’s reputation? Do they still perceive two
distinct reputational cues – manufacturer brand(s) and retailer brand – and if so,
how do they interact in the attitude formation process? Which of the two (very
different) roles dominates their perception? And how does this perception then
influence their evaluation of the firm’s offering? An initial hypothesis is that this
may strongly depend on different types of customer. Some will “buy the brand”,
and show little interest in other parts of the offering. It seems quite plausible that
a certain type of Audi customer could not care less about the fact that the
dealer’s next door showroom sports Volkswagen cars. Another type of client
may just be looking for a mid-size quality car and therefore appreciate the
chance to directly compare the Audi A3 and the VW Golf in form of a “onestop-shopping”. It would therefore seem promising if further research
investigated the different perceived roles of companies that offer an open
architecture, how these perceptions influence customer reactions and how they
relate to customer characteristics and needs.
129
5.5.3 Perceived Product and Range Fit
Finally, a field of potential research emerges from differences not among
customers or firms, but products. Some of the statements from the qualitative
pre-study hint at an interesting effect: In what they regard as ‘standard’
investment categories (e.g., a mutual fund replicating the Dow Jones index), the
financially experienced clients will expect and demand their bank to offer some
of its own products. If this is not the case, some clients tend to attribute this to
the overall bad quality of the bank’s products, questioning its role as a
manufacturer of investment solutions (“If they cannot make that, what can
they?”). These unfavorable attributions do not occur where ‘specialty’
investments are discussed (e.g., a fund that invests in fine art). Here, financially
experienced clients welcome the offering of third-party products, because their
manufacturers promise a superior expertise that the bank itself is not expected to
have. The perceived proximity or distance of a product to its manufacturer’s
core offering has been found influential in other studies, especially on brand
extensions and product fit. They suggest that customers perceive new products
sold under a specific brand more favorably if they share a certain similarity with
the category or categories in which the parent brand traditionally operates
(Aaker and Keller 1990; Boush and Loken 1991; Broniarczyk and Alba 1994;
Buil, de Chernatony, and Hem 2009). Keller and Aaker (1992) argue that one
dimension of such fit evaluations pertains to a firm's perceived ability to
manufacture the extension. From a customer’s view-point, certain products may
‘overstretch’ a firm’s expertise and capabilities (Smith and Andrews 1995). In
the context of the present research, this means: While many customers will, for
instance, acknowledge Bank of America's (BofA) extensive range of products
and global investment expertise, they may be doubtful of its ability to set up a
highly specialized mutual fund on ‘classical European art’. Instead, they may
welcome it if their BofA advisor offered them such a fund from an Amsterdambased fund management ‘boutique’ specialized on art investment. In short, an
open-architecture extension of a firm’s product portfolio may be best accepted
130
by customers if it focused on those products that are dissimilar to the firm’s inhouse offering and require capabilities in which the company can claim little
competence.
Does this on the other hand mean that companies can sell virtually anything, as
long as it is from a third-party supplier and not branded as their own? That
seems quite unlikely. Rather, even the range of third-party products that can be
sold through an open architecture will be limited. Assortment variety research
has shown that consumers also evaluate the overall fit of a retailer's image and a
specific product range (‘range fit’). If a product range is too dissimilar to a
retailer's image, consumers can perceive those products as ‘inappropriate’ for
the retailer (Hart and Davies 1996). Great American Bank, for instance, a
Kansas-based local community bank, may not be perceived to have sufficient
expertise for even selling any investment products that cover, e.g., Asian-Pacific
economies, no matter whether they are own-label or sourced from an external
provider. It can be concluded that, in order for the consumer to appreciate thirdparty products, these products may need to be sufficiently dissimilar to the
manufacturer's traditional offering, and sufficiently similar to its image as a
retailer. Additional research in this area seems promising and relevant – it may
reveal insights into which part of a company’s offering could be enhanced by
adding third-party products and which should not.
131
6
REFERENCES
Aaker, David A. and Kevin Lane Keller (1990), "Consumer Evaluations of
Brand Extensions," Journal of Marketing, 54 (1), 27-41.
Ahearne, Michael, Ronald Jelinek, and Eli Jones (2007), "Examining the Effect
of Salesperson Service Behavior in a Competitive Context," Journal of
the Academy of Marketing Science, 35 (4), 603-16.
Alba, Joseph W. and J. Wesley Hutchinson (1987), "Dimensions of Consumer
Expertise," Journal of Consumer Research, 13 (4), 411-54.
Aldlaigan, Abdullah H. and Francis A. Buttle (2001), "Consumer Involvement
in Financial Services: An Empirical Test of Two Measures," International
Journal of Bank Marketing, 19 (6), 232.
Amaldoss, Wilfred, Robert J. Meyer, Jagmohan S. Raju, and Amnon Rapoport
(2000), "Collaborating to Compete," Marketing Science, 19 (2), 105-26.
Anderson, Eugene W. and Mary W. Sullivan (1993), "The Antecedents and
Consequences of Customer Satisfaction for Firms," Marketing Science, 12
(2), 125-43.
Anonymous (2006a), "Adig-Fonds Werden in Cominvest Umbenannt,"
www.fondscheck.de, September 26, 2006, Accessed: April 20, 2011,
http://www.fondscheck.de/artikel/news-Sonstiges-1376124.html.
Anonymous (2006b), "Commerzbank Stampft Die Marke Adig Ein,"
www.handelsblatt.com, Accessed: April 20, 2011,
http://www.handelsblatt.com/finanzen/fonds/nachrichten/commerzbankstampft-die-marke-adig-ein/2657402.html.
Anonymous (2010), "Private Banking Units Staff Up," in Euromoney.
Anselmi, Kenneth and James E. Zemanek Jr (1997), "Relationship Selling: How
Personal Characteristics of Salespeople Affect Buyer Satisfaction,"
Journal of Social Behavior & Personality, 12 (2), 539-50.
Areni, Charles S. and K. Chris Cox (1995), "Assessing the Impact of Message
Cues and Arguments in Persuasion: Conceptual and Methodological
Issues," Advances in Consumer Research, 22 (1), 198-202.
132
Baker, Julie, A. Parasuraman, Dhruv Grewal, and Glenn B. Voss (2002), "The
Influence of Multiple Store Environment Cues on Perceived Merchandise
Value and Patronage Intentions," Journal of Marketing, 66 (2), 120-41.
Baron, Reuben M. and David A. Kenny (1986), "The Moderator–Mediator
Variable Distinction in Social Psychological Research: Conceptual,
Strategic, and Statistical Considerations," Journal of Personality and
Social Psychology, 51 (6), 1173-82.
Barr, N. (2009), "Marks and Spencer to Start Selling Branded Food Products.,"
http://www.express.co.uk, November 4, 2009, Accessed: November 14,
2009, http://www.express.co.uk/posts/view/138154/Marks-Spencer-tostart-selling-branded-food-products.
Bateson, John E. G. and Michael K. Hui (1992), "The Ecological Validity of
Photographic Slides and Videotapes in Simulating the Service Setting,"
Journal of Consumer Research, 19 (2), 271-81.
Baum, Klaus-Jürgen (2005), "Institute Öffnen Sich Fremden Fonds,"
Handelsblatt, April 6, b08.
Bell, Simon J., Seigyoung Auh, and Karen Smalley (2005), "Customer
Relationship Dynamics: Service Quality and Customer Loyalty in the
Context of Varying Levels of Customer Expertise and Switching Costs,"
Journal of the Academy of Marketing Science, 33 (2), 169-83.
Bell, Simon J. and Andreas B. Eisingerich (2007), "The Paradox of Customer
Education: Customer Expertise and Loyalty in the Financial Services
Industry," European Journal of Marketing, 41 (5/6), 466-86.
Bender, Yuri (2009), "European Banks Cut Access to External Fund Providers,"
www.ft.com, September 11, 2010, Accessed: November 28, 2010,
http://www.ft.com/cms/s/0/76abe988-9979-11de-ab8c00144feabdc0.html?nclick_check=1.
Berger, Ida E., Peggy H. Cunningham, and Robert V. Kozinets (1999),
"Consumer Persuasion through Cause-Related Advertising," Advances in
Consumer Research, 26 (1), 491-97.
133
Beverland, Michael, Francis Farrelly, and Zeb Woodhatch (2007), "Exploring
the Dimensions of Proactivity within Advertising Agency--Client
Relationships," Journal of Advertising, 36 (4), 49-60.
Bitner, Mary Jo (1990), "Evaluating Service Encounters: The Effects of Physical
Surroundings and Employee Responses," Journal of Marketing, 54 (2),
69-82.
Bloemer, Josée, Ko de Ruyter, and Martin Wetzels (1999), "Linking Perceived
Service Quality and Service Loyalty: A Multi-Dimensional Perspective,"
European Journal of Marketing, 33 (11/12), 1082-106.
Boatwright, Peter and Joseph C. Nunes (2001), "Reducing Assortment: An
Attribute-Based Approach," Journal of Marketing, 65 (3), 50-63.
Boles, James S., Julie T. Johnson, and Hiram C. Barksdale Jr (2000), "How
Salespeople Build Quality Relationships: A Replication and Extension,"
Journal of Business Research, 48 (1), 75-81.
Bolton, Patrick, Xavier Freixas, and Joel Shapiro (2007), "Conflicts of Interest,
Information Provision, and Competition in the Financial Services
Industry," Journal of Financial Economics, 85 (2), 297-330.
Bolton, Ruth N. and Katherine N. Lemon (1999), "A Dynamic Model of
Customers' Usage of Services: Usage as an Antecedent and Consequence
of Satisfaction," Journal of Marketing Research (JMR), 36 (2), 171-86.
Boorom, Michael L., Jerry R. Goolsby, and Rosemary P. Ramsey (1998),
"Relational Communication Traits and Their Effect on Adaptiveness and
Sales Performance," Journal of the Academy of Marketing Science, 26 (1),
16-30.
Boulding, William and Amna Kirmani (1993), "A Consumer-Side Experimental
Examination of Signaling Theory: Do Consumers Perceive Warranties as
Signals of Quality?," The Journal of Consumer Research, 20 (1), 111-23.
Boush, David M. and Barbara Loken (1991), "A Process-Tracing Study of
Brand Extension Evaluation," Journal of Marketing Research (JMR), 28
(1), 16-28.
134
Broniarczyk, Susan M. and Joseph W. Alba (1994), "The Importance of the
Brand in Brand Extension," Journal of Marketing Research (JMR), 31 (2),
214-28.
Broniarczyk, Susan M., Wayne D. Hoyer, and Leigh McAlister (1998),
"Consumers' Perceptions of the Assortment Offered in a Grocery
Category: The Impact of Item Reduction," Journal of Marketing Research
(JMR), 35 (2), 166-76.
Brown, Tom J., Thomas E. Barry, Peter A. Dacin, and Richard F. Gunst (2005),
"Spreading the Word: Investigating Antecedents of Consumers' Positive
Word-of-Mouth Intentions and Behaviors in a Retailing Context," Journal
of the Academy of Marketing Science, 33 (2), 123-38.
Buil, Isabel, Leslie de Chernatony, and Leif E. Hem (2009), "Brand Extension
Strategies: Perceived Fit, Brand Type, and Culture Influences," European
Journal of Marketing, 43 (11/12), 1300-24.
Burger, Jerry M. and Lawton T. Hemans (1988), "Desire for Control and the
Use of Attribution Processes," Journal of Personality, 56 (3), 531-46.
Burnkrant, R. E. (1975), "Attribution Theory in Marketing Research: Problems
and Prospects," Advances in Consumer Research, 2 (1), 465.
Campbell, Margaret C. and Amna Kirmani (2000), "Consumers' Use of
Persuasion Knowledge: The Effects of Accessibility and Cognitive
Capacity on Perceptions of an Influence Agent," Journal of Consumer
Research, 27 (1), 69-83.
Casado Diaz, Ana B. and Francisco J. Más Ruíz (2002), "The Consumer's
Reaction to Delays in Service," International Journal of Service Industry
Management, 13 (2), 118.
Challagalla, Goutam, R. Venkatesh, and Ajay K. Kohli (2009), "Proactive
Postsales Service: When and Why Does It Pay Off?," Journal of
Marketing, 73 (2), 70-87.
Chebat, Jean-Charles, Pierre Filiatrault, Claire Gélinas-Chebat, and Alexander
Vaninsky (1995), "Impact of Waiting Attribution and Consumer's Mood
on Perceived Quality," Journal of Business Research, 34 (3), 191-96.
135
Chernev, Alexander (2003), "Product Assortment and Individual Decision
Processes," Journal of Personality & Social Psychology, 85 (1), 151-62.
Chernev, Alexander (2006), "Decision Focus and Consumer Choice among
Assortments," Journal of Consumer Research, 33 (1), 50-59.
Chernev, Alexander and Leigh McAllister (2005), "Product Assortment and
Variety-Seeking in Consumer Choice," Advances in Consumer Research,
32 (1), 119-21.
Crosby, Lawrence A., Kenneth A. Evans, and Deborah Cowles (1990),
"Relationship Quality in Services Selling: An Interpersonal Influence
Perspective," Journal of Marketing, 54 (3), 68.
Crowley, Ayn E. and Wayne D. Hoyer (1994), "An Integrative Framework for
Understanding Two-Sided Persuasion," Journal of Consumer Research,
20 (4), 561-74.
Cunningham, John D. and Harold H. Kelley (1975), "Causal Attributions for
Interpersonal Events of Varying Magnitude1," Journal of Personality, 43
(1), 74-93.
Curren, Mary T. and Valerie S. Folkes (1987), "Attributional Influences on
Consumers' Desires to Communicate About Products," Psychology &
Marketing, 4 (1), 31-45.
Darke, Peter R. and Robin J. B. Ritchie (2007), "The Defensive Consumer:
Advertising Deception, Defensive Processing, and Distrust," Journal of
Marketing Research (JMR), 44 (1), 114-27.
de Bont, Cees J. P. M. and Jan P. L. Schoormans (1995), "The Effects of
Product Expertise on Consumer Evaluations of New-Product Concepts,"
Journal of Economic Psychology, 16 (4), 599.
de Matos, Celso Augusto and Carlos Alberto Vargas Rossi (2008), "Word-ofMouth Communications in Marketing: A Meta-Analytic Review of the
Antecedents and Moderators," Journal of the Academy of Marketing
Science, 36 (4), 578-96.
136
DeCarlo, Thomas E. (2005), "The Effects of Sales Message and Suspicion of
Ulterior Motives on Salesperson Evaluation," Journal of Consumer
Psychology (Lawrence Erlbaum Associates), 15 (3), 238-49.
Dion, Paul A. and Elaine M. Notarantonio (1992), "Salesperson Communication
Style: The Neglected Dimension in Sales Performance," Journal of
Business Communication, 29 (1), 63-77.
Dixon, Andrea L., Rosann L. Spiro, and Maqbul Jamil (2001), "Successful and
Unsuccessful Sales Calls: Measuring Salesperson Attributions and
Behavioral Intentions," Journal of Marketing, 65 (3), 64-78.
Dodds, William B., Kent B. Monroe, and Dhruv Grewal (1991), "Effects of
Price, Brand, and Store Information on Buyers' Product Evaluations,"
Journal of Marketing Research (JMR), 28 (3), 307-19.
Doney, Patricia M. and Joseph P. Cannon (1997), "An Examination of the
Nature of Trust in Buyer-Seller Relationships," Journal of Marketing, 61
(2), 35.
Dubinsky, Alan J., Steven J. Skinner, and Tommy E. Whittler (1989),
"Evaluating Sales Personnel: An Attribution Theory Perspective," Journal
of Personal Selling & Sales Management, 9 (1), 9.
Dussauge, Pierre and Bernard Garrette (1999), Cooperative Strategy :
Competing Successfully through Strategic Alliances, New York: John
Wiley.
Dussauge, Pierre, Bernard Garrette, and Will Mitchell (2004), "Asymmetric
Performance: The Market Share Impact of Scale and Link Alliances in the
Global Auto Industry," Strategic Management Journal, 25 (7), 701-11.
Eagly, Alice H., Wendy Wood, and Shelly Chaiken (1978), "Causal Inferences
About Communicators and Their Effect on Opinion Change," Journal of
Personality and Social Psychology, 36 (4), 424-35.
Eisend, Martin (2007), "Understanding Two-Sided Persuasion: An Empirical
Assessment of Theoretical Approaches," Psychology and Marketing, 24
(7), 615-40.
137
Elig, Timothy W. and Irene H. Frieze (1979), "Measuring Causal Attibutions for
Success and Failure," Journal of Personality and Social Psychology, 37
(4), 621-34.
Enzle, Michael E. and Donald Schopflocher (1978), "Instigation of Attribution
Processes by Attributional Questions," Personality and Social Psychology
Bulletin, 4 (4), 595-99.
Etgar, Michael and Stephen A. Goodwin (1982), "One-Sided Versus Two-Sided
Comparative Message Appeals for New Brand Introductions," Journal of
Fein, Steven (1996), "Effects of Suspicion on Attributional Thinking and the
Correspondence Bias," Journal of Personality & Social Psychology, 70
(6), 1164-84.
Felsted, Andrea (2009), "M&S to Expand Brands Sales in Grocery Wars,"
www.FT.com, November 4, 2009, Accessed: November 16, 2009,
http://proquest.umi.com/pqdweb?did=1893831871&Fmt=7&clientId=456
08&RQT=309&VName=PQD.
Finch, Julia (2009), "This Is Not Just M&S Food. Store Gives Shelf Space to
Top Brands," www.guardian.co.uk, November 4, 2009, Accessed:
November 4, 2009, http://www.guardian.co.uk/business/2009/
nov/04/marks-and-spencer-food.
Fletcher, Garth J. (1984), "Psychology and Common Sense," American
Psychologist, 39 (3), 203-13.
Folkes, Valerie S. (1984), "Consumer Reactions to Product Failure: An
Attributional Approach," Journal of Consumer Research, 10 (4), 398-409.
Folkes, Valerie S. (1988), "Recent Attribution Research in Consumer Behavior:
A Review and New Directions," Journal of Consumer Research, 14 (4),
548-65.
Folkes, Valerie S., Susan Koletsky, and John L. Graham (1987), "A Field Study
of Causal Inferences and Consumer Reaction: The View from the
Airport," Journal of Consumer Research, 13 (4), 534-39.
138
Friestad, Marian and Peter Wright (1994), "The Persuasion Knowledge Model:
How People Cope with Persuasion Attempts," Journal of Consumer
Research, 21 (1), 1-31.
Garella, Paolo G. and Martin Peitz (2007), "Alliances between Competitors and
Consumer Information," Journal of the European Economic Association,
5 (4), 823-45.
Gimbel, Florian and Tony Major (2002), "Germany Flocks to One-Stop Shops:
Tony Major and Florian Gimbel on the Banking Trend Towards 'Open
Architecture'," Financial Times, September 16, 07.
Goff, Brent G., James S. Boles, Danny N. Bellenger, and Carrie Stojack (1997),
"The Influence of Salesperson Selling Behaviors on Customer Satisfaction
with Products," Journal of Retailing, 73 (2), 171-83.
Golden, Linda L. (1977), "Attribution Theory Implications for Advertisement
Claim Credibility," Journal of Marketing Research (JMR), 14 (1), 115-17.
Golden, Linda L. and Mark I. Alpert (1987), "Comparative Analysis of the
Relative Effectiveness of One- and Two-Sided Communication for
Contrasting Products," Journal of Advertising, 16 (1), 18-68.
GoldmanSachs (2011), "Print Advertisement," in Harvard Business Review,
Vol. 89, 2-3.
Grandey, Alicia A., Glenda M. Fisk, Anna S. Mattila, Karen J. Jansen, and Lori
A. Sideman (2005), "Is "Service with a Smile" Enough? Authenticity of
Positive Displays During Service Encounters," Organizational Behavior
& Human Decision Processes, 96 (1), 38-55.
Greenleaf, Eric A. and Donald R. Lehmann (1995), "Reasons for Substantial
Delay in Consumer Decision Making," Journal of Consumer Research, 22
(2), 186-99.
Greenwald, Anthony G. and Clark Leavitt (1984), "Audience Involvement in
Advertising: Four Levels," Journal of Consumer Research, 11 (1), 58192.
Grewal, Dhruv, R. Krishnan, Julie Baker, and Norm Borin (1998), "The Effects
of Store Name, Brand Name and Price Discounts on Consumers'
139
Evaluations and Purchase Intentions," Journal of Retailing, 74 (3), 33152.
Hagedoorn, John, Albert N. Link, and Nicholas S. Vonortas (2000), "Research
Partnerships," Research Policy, 29 (4-5), 567-86.
Hamel, Gary, Yves L. Doz, and C. K. Prahalad (1989), "Collaborate with Your
Competitors--and Win," Harvard Business Review, 67 (1), 133-39.
Hart, Cathy A. and Mark A. P. Davies (1996), "Consumer Perceptions of NonFood Assortments: An Empirical Study," Journal of Marketing
Management, 12 (4), 297-312.
Hastie, Reid (1984), "Causes and Effects of Causal Attribution," Journal of
Hawes, Jon M., Kenneth E. Mast, and John E. Swan (1989), "Trust Earning
Perceptions of Sellers and Buyers," Journal of Personal Selling & Sales
Heider, Fritz (1958), The Psychology of Interpersonal Relations, New York:
Wiley.
Herkner, Werner (1981), Einführung in Die Sozialpsychologie, Bern, Stuttgart,
Wien: Verlag Hans Huber.
Hess Jr, Ronald L., Shankar Ganesan, and Noreen M. Klein (2003), "Service
Failure and Recovery: The Impact of Relationship Factors on Customer
Satisfaction," Journal of the Academy of Marketing Science, 31 (2), 12745.
Hoch, Stephen J., Eric T. Bradlow, and Brian Wansink (1999), "The Variety of
an Assortment," Marketing Science, 18 (4), 527-46.
Howcroft, Barry, Robert Hamilton, and Paul Hewer (2007), "Customer
Involvement and Interaction in Retail Banking: An Examination of Risk
and Confidence in the Purchase of Financial Products," Journal of
Services Marketing, 21 (7), 481-91.
Huffman, Cynthia and Barbara E. Kahn (1998), "Variety for Sale: Mass
Customization or Mass Confusion?," Journal of Retailing, 74 (4), 491513.
140
Hui, Michael K. and Roy Toffoli (2002), "Perceived Control and Consumer
Attribution for the Service Encounter," Journal of Applied Social
Psychology, 32 (9), 1825-44.
Hui, Michael K., Xiande Zhao, Xiuncheng Fan, and Kevin Au (2004), "When
Does the Service Process Matter? A Test of Two Competing Theories,"
Hunt, James M., Teresa J. Domzal, and Jerome B. Kernan (1982), "Causal
Attributions and Persuasion: The Case of Disconfirmed Expectancies,"
Advances in Consumer Research, 9 (1), 287-92.
Hunt, James M., Michael F. Smith, and Jerome B. Kernan (1985), "The Effects
of Expectancy Disconfirmation and Argument Strength on Message
Processing Level: An Application to Personal Selling," Advances in
HypoVereinsbank (2001), "Print Advertisement," Munich: Historical Archive of
UniCredit Bank AG.
Iglesias, Victor (2009), "The Attribution of Service Failures: Effects on
Consumer Satisfaction," The Service Industries Journal, 29 (2), 127-41.
Iyengar, Sheena S. and Mark R. Lepper (2000), "When Choice Is Demotivating:
Can One Desire Too Much of a Good Thing?," Journal of Personality &
Social Psychology, 79 (6), 995-1006.
Jacoby, Jacob and David Mazursky (1984), "Research Note: Linking Brand and
Retailer Images--Do the Potential Risks Outweigh the Potential
Benefits?," Journal of Retailing, 60 (2), 105.
Johnson, Mark S. (2006), "A Bibliometric Review of the Contribution of
Attribution Theory to Sales Management," Journal of Personal Selling &
Sales Management, 26 (2), 181-95.
Johnson, Steve (2011), "Bank-Run Funds Are Poor Performers," www.ft.com,
January 9, 2011, Accessed: January 23, 2011, http://proquest.umi.com/
pqdlink?did=2234118501&Fmt=7&clientId=45608&RQT=309&VName
=PQD.
141
Jonas, Eva and Dieter Frey (2003), "Information Search and Presentation in
Advisor–Client Interactions," Organizational Behavior & Human
Decision Processes, 91 (2), 154.
Jonas, Eva, Stefan Schulz-Hardt, Dieter Frey, and Norman Thelen (2001),
"Confirmation Bias in Sequential Information Search after Preliminary
Decisions: An Expansion of Dissonance Theoretical Research on
Selective Exposure to Information," Journal of Personality & Social
Psychology, 80 (4), 557-71.
Jones, Edward E. and Keith Davis (1965), "From Acts to Dispositions: The
Attribution Process in Person Perception," in Advances in Experimental
Social Psychology, Vol. 2, ed. Leonard Berkowitz, New York: Academic
Press, 219-66.
Kahn, Barbara E. (1995), "Consumer Variety-Seeking among Goods and
Services : An Integrative Review," Journal of Retailing and Consumer
Services, 2 (3), 139-48.
Kahn, Barbara E. and Donald R. Lehmann (1991), "Modeling Choice among
Assortments," Journal of Retailing, 67 (3), 274.
Kelleher, Ellen (2007), "Objective Advice - and Widest Range of Products Open
Architecture: Outside Managers Are Key to Banks' Strategies, Writes
Ellen Kelleher," Financial Times, June 23, 17.
Keller, Kevin Lane and David A. Aaker (1992), "The Effects of Sequential
Introduction of Brand Extensions," Journal of Marketing Research (JMR),
29 (1), 35-60.
Keller, Kevin Lane and Donald R. Lehmann (2006), "Brands and Branding:
Research Findings and Future Priorities," Marketing Science, 25 (6), 74059.
Kelley, Harold H. (1967), "Attribution Theory in Social Psychology," in
Nebraska Symposium on Motivation, ed. David Levine, Lincoln, NE:
University of Nebraska Press, 192-238.
Kelley, Harold H. (1973), "The Process of Causal Attribution," American
Psychologist (February), 21.
142
Kelley, Harold H. and John L. Michela (1980), "Attribution Theory and
Research," Annual Review of Psychology, 31, 457-501.
Kennedy, Mary Susan, Linda K. Ferrell, and Debbie Thorne LeClair (2001),
"Consumers' Trust of Salesperson and Manufacturer: An Empirical
Study," Journal of Business Research, 51 (1), 73-86.
Kirmani, Amna and Margaret C. Campbell (2004), "Goal Seeker and Persuasion
Sentry: How Consumer Targets Respond to Interpersonal Marketing
Persuasion," Journal of Consumer Research, 31 (3), 573-82.
Koeske, Gary F. and William D. Crano (1968), "The Effect of Congruous and
Incongruous Source-Statement Combinations Upon the Judged Credibility
of a Communication," Journal of Experimental Social Psychology, 4 (4),
384-99.
Kohn, Paul M. and Suzi Snook (1976), "Expectancy--Violation, Similarity, and
Unexpected Similarity as Sources of Credibility and Persuasiveness,"
Journal of Psychology, 94 (2), 185.
Krausz, Miriam and Jacob Paroush (2002), "Financial Advising in the Presence
of Conflict of Interests," Journal of Economics & Business, 54 (1), 55.
Kroeber-Riel, Werner (2003), Konsumentenverhalten, München: Kroeber-Riel,
Werner, Weinberg, Peter.
Lau, Richard R. and Dan Russell (1980), "Attributions in the Sports Pages,"
Journal of Personality and Social Psychology, 39 (1), 29-38.
Lichtenstein, Donald R. and William O. Bearden (1986), "Measurement and
Structure of Kelley's Covariance Theory," Journal of Consumer Research,
13 (2), 290-96.
Lichtenstein, Donald R. and Scot Burton (1988), "The Measurement and
Moderating Role of Confidence in Attributions," Advances in Consumer
Research, 15 (1), 468-75.
Luo, Xueming, Aric Rindfleisch, and David K. Tse (2007), "Working with
Rivals: The Impact of Competitor Alliances on Financial Performance,"
143
Maxham III, James G. and Richard G. Netemeyer (2002), "Modeling Customer
Perceptions of Complaint Handling over Time: The Effects of Perceived
Justice on Satisfaction and Intent," Journal of Retailing, 78 (4), 239.
McPeek, Robert W. and John D. Edwards (1975), "Expectancy Disconfirmation
and Attitude Change," Journal of Social Psychology, 96 (2), 193.
Michaels, Ronald E. and Ralph L. Day (1985), "Measuring Customer
Orientation of Salespeople: A Replication with Industrial Buyers,"
Mizerski, Richard W., Linda L. Golden, and Jerome B. Kernan (1979), "The
Attribution Process in Consumer Decision Making," Journal of Consumer
Research, 6 (2), 123-40.
Mulhern, Francis J. (1997), "Retail.Marketing: From Distribution to
Integration," International Journal of Research in Marketing, 14 (2), 10324.
Narat, Info (2002), "Teure Fonds-Demokratie," Handelsblatt, May 6, 29.
Niebuhr, Robert E., Charles C. Manz, and Kermit R. Davis Jr (1981), "Using
Videotape Technology: Innovations in Behavioral Research," Journal of
Management, 7 (2), 43-54.
Niemeyer, Hans-Georg (1993), Begründungsmuster Von Konsumenten :
Attributionstheoretische Grundlagen Und Einflussmöglichkeiten Im
Marketing, Heidelberg: Physica-Verlag.
Oppewal, Harmen and Kitty Koelemeijer (2005), "More Choice Is Better:
Effects of Assortment Size and Composition on Assortment Evaluation,"
International Journal of Research in Marketing, 22 (1), 45-60.
Pechmann, Cornelia (1992), "Predicting When Two-Sided Ads Will Be More
Effective Than One Sided Ads: The Role of Correlation and
Correspondent Inferences," Journal of Marketing Research (JMR), 29 (4),
441-53.
Pfanner, Eric (2002), "Broadening Horizons When Times Are Tough,"
International Herald Tribune, October 31, 6.
144
Plouffe, Christopher R., John Hulland, and Trent Wachner (2009), "CustomerDirected Selling Behaviors and Performance: A Comparison of Existing
Perspectives," Journal of the Academy of Marketing Science, 37 (4), 42239.
Pornpitakpan, Chanthika (2004), "The Persuasiveness of Source Credibility: A
Critical Review of Five Decades' Evidence," Journal of Applied Social
Psychology, 34 (2), 243-81.
Porter, Stephen S. and Cindy Claycomb (1997), "The Influence of Brand
Recognition on Retail Store Image," The Journal of Product and Brand
Purohit, Devavrat and Joydeep Srivastava (2001), "Effect of Manufacturer
Reputation, Retailer Reputation, and Product Warranty on Consumer
Judgments of Product Quality: A Cue Diagnosticity Framework," Journal
of Consumer Psychology (Lawrence Erlbaum Associates), 10 (3), 123-34.
Pyszczynski, Thomas A. and Jeff Greenberg (1981), "Role of Disconfirmed
Expectancies in the Instigation of Attributional Processing," Journal of
Rajaobelina, Lova and Jasmin Bergeron (2009), "Antecedents and
Consequences of Buyer-Seller Relationship Quality in the Financial
Services Industry," International Journal of Bank Marketing, 27 (5), 35980.
Rao, Akshay R. and Kent B. Monroe (1988), "The Moderating Effect of Prior
Knowledge on Cue Utilization in Product Evaluations," Journal of
Rasch, Michael (2003), "Viele Fragezeichen Beim Drittfonds-Vertrieb," Neue
Zürcher Zeitung, July 25, 27.
Rentz, Joseph O., C. David Shepherd, Armen Tashchian, Pratibha A. Dabholkar,
and Robert T. Ladd (2002), "A Measure of Selling Skill: Scale
Development and Validation," Journal of Personal Selling & Sales
Management, 22 (1), 13-21.
145
Richard, Buda and Zhang Yong (2000), "Consumer Product Evaluation: The
Interactive Effect of Message Framing, Presentation Order, and Source
Credibility," The Journal of Product and Brand Management, 9 (4), 229.
Rose, Randall L. and Peter R. Dickson (1987), ""He Says No, but Does He
Really Mean It?": Bargaining Behavior, Cue Consistency, and
Attribution," Advances in Consumer Research, 14 (1), 382-86.
Rose, Randall L. and Peter R. Dickson (1988), "An Initial Test of the Effects of
Cue Patterns on Behavior and Attributions in a Purchasing Negotiation,"
Ross, Alice (2010), ""Own-Brand" Products Fly Off the Shelves," Financial
Times, July 3, 6-8.
Rust, Roland T. and Anthony J. Zahorik (1993), "Customer Satisfaction,
Customer Retention, and Market Share," Journal of Retailing, 69 (2), 193215.
Sanbonmatsu, David M. and Frank R. Kardes (1988), "The Effects of
Physiological Arousal on Information Processing and Persuasion,"
Sandler, Kathy (2009), "M&S Will Add Name-Brand Goods – Retailer Ending
50 Years of Selling Only Its Brands in Nod to Consumer Demands," The
Wallstreet Journal Europe, November 5, 8.
Saxe, Robert and Barton A. Weitz (1982), "The Soco Scale: A Measure of the
Customer Orientation of Salespeople," Journal of Marketing Research
(JMR), 19 (3), 343-51.
Schulz, Thomas M. (2002), "Anzeichen Eines Paradigmenwechsels in Der
Fondsproduktion," in Zeitschrift für das gesamte Kreditwesen.
Severin, Christin (2002), "Öffnung Für Den Vertrieb Von Drittfonds: Nicht
Immer Meinen Es Die Banken Wirklich Ernst," Neuer Zürcher Zeitung,
January 29, 71.
Sharma, Arun (1996), "The Effect of Priming Cues in Sales Interactions:
Additional Perspectives," Journal of Personal Selling & Sales
Management, 16 (2), 49-52.
146
Sharma, Neeru and Paul G. Patterson (1999), "The Impact of Communication
Effectiveness and Service Quality on Relationship Commitment In,"
Journal of Services Marketing, 13 (2/3), 151.
Sharma, Neeru and Paul G. Patterson (2000), "Switching Costs, Alternative
Attractiveness and Experience as Moderators of Relationship
Commitment in Professional, Consumer Services," International Journal
of Service Industry Management, 11 (5), 470.
Shaughnessy, Gary (2009), "Don't Kill Off Open Architecture and Choice,"
Financial Times, November 30, 6.
Simonson, Itamar (1990), "The Effect of Purchase Quantity and Timing on
Variety-Seeking Behavior," Journal of Marketing Research (JMR), 27
(2), 150-62.
Sirdeshmukh, Deepak, Jagdip Singh, and Barry Sabol (2002), "Consumer Trust,
Value, and Loyalty in Relational Exchanges," Journal of Marketing, 66
(1), 15-37.
Skinner, Mark (2006), "Choice Galore with Open Access to Best Funds Open
Architecture Has Enabled Banks to Tap into Demand for Actively
Managed Funds, Accessing Europe's Most Talented Managers, Says Mark
Skinner," Financial Times, March 27, 6.
Smith, Michael F. and James M. Hunt (1987), "Disconfirmation of
Expectations: A Method for Enhancing the Effectiveness of Customer
Communications," Journal of Personal Selling & Sales Management, 7
(1), 9.
Sparks, John R. and Charles S. Areni (2002), "The Effects of Sales Presentation
Quality and Initial Perceptions on Persuasion: Multiple Role Perspective,"
Journal of Business Research, 55 (6), 517-28.
Speck, Kurt (2010), "Anlagefonds Sind Gewinner Der Krise," HandelsZeitung,
February 3, 41.
Stafford, Thomas F. (1996), "Conscious and Unconscious Processing of Priming
Cues in Selling Encounters," Journal of Personal Selling & Sales
Management, 16 (2), 37-44.
147
Stafford, Thomas F., Thomas W. Leigh, and Leonard L. Martin (1995),
"Assimilation and Contrast Priming Effects in the Initial Consumer Sales
Call," Psychology & Marketing, 12 (4), 321-47.
Su, Wanru and Michael J. Tippins (1998), "Consumer Attributions of Product
Failure to Channel Members and Self: The Impacts of Situational Cues,"
Sujan, Mita, James R. Bettman, and Harish Sujan (1986), "Effects of Consumer
Expectations on Information Processing in Selling Encounters," Journal
of Marketing Research (JMR), 23 (4), 346-53.
Swan, John E., Michael R. Bowers, and Lynne D. Richardson (1999),
"Customer Trust in the Salesperson: An Integrative Review and MetaAnalysis of the Empirical Literature," Journal of Business Research, 44
(2), 93-107.
Swan, John E. and Richard L. Oliver (1989), "Postpurchase Communications by
Consumers," Journal of Retailing, 65 (4), 516.
Swanson, Scott R. and Scott W. Kelley (2001), "Attributions and Outcomes of
the Service Recovery Process," Journal of Marketing Theory & Practice,
9 (4), 50.
Tom, Gail and Scott Lucey (1995), "Waiting Time Delays and Customer
Satisfaction in Supermarkets," Journal of Services Marketing, 9 (5), 20.
Tormala, Zakary L. and Richard E. Petty (2004), "Source Credibility and
Attitude Certainty: A Metacognitive Analysis of Resistance to
Persuasion," Journal of Consumer Psychology (Lawrence Erlbaum
Associates), 14 (4), 427-42.
Trommsdorff, Volker (2009), Konsumentenverhalten, Stuttgart: Kohlhammer.
Tsiros, Michael, Vikas Mittal, and Willaim T. Ross Jr (2004), "The Role of
Attributions in Customer Satisfaction: A Reexamination," Journal of
Weber, Katy (2002), "Aufgerissene Pforten Für Fremdfonds," www.managermagazin.de, March 10, 2002, Accessed: October 9, 2008,
148
http://www.manager-magazin.de/unternehmen/artikel/
0,2828,192442,00.html.
Weiner, Bernard (1985a), "An Attributional Theory of Achievement Motivation
and Emotion," Psychological Review, 92 (4), 548-73.
Weiner, Bernard (1985b), ""Spontaneous" Causal Thinking," Psychological
Bulletin, 97 (1), 74-84.
Weitz, Barton A. (1981), "Effectiveness in Sales Interactions: A Contingency
Framework," Journal of Marketing, 45 (1), 85-103.
Weitz, Barton A., Harish Sujan, and Mita Sujan (1986), "Knowledge,
Motivation, and Adaptive Behavior: A Framework for Improving Selling
Effectiveness," Journal of Marketing, 50 (4), 174-91.
Wentzel, Daniel (2009), "The Effect of Employee Behavior on Brand
Personality Impressions and Brand Attitudes," Journal of the Academy of
Marketing Science, 37 (3), 359-74.
Wentzel, Daniel, Torsten Tomczak, and Andreas Herrmann (2008), "Wirkung
Des Mitarbeiterverhaltens Auf Die Markenpersönlichkeit," Marketing
ZFP (3), 133.
White, Tiffany Barnett (2005), "Consumer Trust and Advice Acceptance:The
Moderating Roles of Benevolence,Expertise, and Negative Emotions,"
Journal of Consumer Psychology (Lawrence Erlbaum Associates), 15 (2),
141-48.
Wiecking, Kai (2003), "Geheimnis À La Fort Knox " www.managermagazin.de, December 4, 2003, Accessed: January 15, 2011,
http://www.manager-magazin.de/finanzen/geldanlage/
0,2828,276868,00.html.
Wimer, Scott and Harold H. Kelley (1982), "An Investigation of the Dimensions
of Causal Attribution," Journal of Personality and Social Psychology, 43
(6), 1142-62.
Wong, Paul T. and Bernard Weiner (1981), "When People Ask "Why"
Questions, and the Heuristics of Attributional Search," Journal of
149
Wood, John Andy, James S. Boles, and Barry J. Babin (2008a), "The Formation
of Buyer's Trust of the Seller in an Initial Sales Encounter," Journal of
Marketing Theory & Practice, 16 (1), 27-39.
Wood, John Andy, James S. Boles, Wesley Johnston, and Danny Bellenger
(2008b), "Buyers' Trust of the Salesperson: An Item-Level MetaAnalysis," Journal of Personal Selling & Sales Management, 28 (3), 26383.
Worsfold, Kate, Jennifer Worsfold, and Graham Bradley (2007), "Interactive
Effects of Proactive and Reactive Service Recovery Strategies: The Case
of Rapport and Compensation," Journal of Applied Social Psychology, 37
(11), 2496-517.
Zaichkowsky, Judith Lynne (1985), "Measuring the Involvement Construct,"
Zeithaml, Valarie A. (2000), "Service Quality, Profitability, and the Economic
Worth of Customers: What We Know and What We Need to Learn,"
Journal of the Academy of Marketing Science, 28 (1), 67-85.
150
7
APPENDICES
7.1 Stimulus Materials Used in Experiment 1
a. Introductory text
Recently, you've decided to benefit from soaring equity prices on the German
and Swiss stock markets. In order to do this, you've put aside a certain amount
of money that you'd like to invest into suitable mutual funds. It's important to
you that you take a well thought-through, sound investment decision. Therefore,
you have made an appointment with the bank that you also use for your other
financial transactions. A while ago, you've read that some banks offer a socalled "open product architecture". Clients of these firms can not only buy inhouse funds managed by the bank itself, but they can choose from a wider
offering of funds that includes external products from third parties. You do not
know whether your bank offers such third-party products, but you intend to find
out.
You meet with Thomas Breiter, your client advisor, in the lobby of the bank's
local branch. After a short welcome, Mr. Breiter accompanies you to his office
where you take seat at a small conference table. After his assistant has brought
you an espresso, you and Mr. Breiter discuss your financial requirements and
expectations for a while. Then, the client advisor suggests to present you with a
selection of products that are suitable for your goals.
b. High proactiveness, balanced product mix, high persuasiveness
Mr. Breiter rises from his chair, walks over to his PC and begins to call up a
variety of fund profiles. He turns around to you and says: "Maybe you've heard
that our firm has an 'open product architecture'? As you nod, he continues: "That
means you don't get only in-house funds from us, but also those of other
151
providers. I'd suggest that I'll also include those third-party products in my
selection."
With a focused expression, the client advisor's glance sweeps across the wide
range of different funds that his PC is displaying. Finally, he prints 1-page
profiles for some of the funds and takes these to the conference table, where he
spreads the pages out in front of you. You leaf through the six documents. Three
of the selected funds are in-house products of the advisor's bank, the other three
are from different external fund managers.
You take a sip of your espresso and give the different fund profiles another
glance. "Ok", you go and lean back in your chair, "why do you recommend
these specific funds?" "Of course", Mr. Breiter smiles, "let me explain that to
you." He than gives you a very accurate and transparent explanation on why he's
chosen each product and what their individual pros and cons are. He also
illustrates very clearly, how these different funds will offer you a good
diversification of your risk.
c. High proactiveness, balanced product mix, low persuasiveness
selection."
152
these specific funds?" "Oh, well" says Mr. Breiter and shrugs. "It's pretty
difficult to explain this in detail". He goes in to long-winded elaboration, using a
lot of jargon. Mr. Breiter does not explain the different pros and cons of each
product, but assures that he's got a "good feeling" about the selection he has
recommended.
d. High proactiveness, biased product mix, high persuasiveness
selection."
spreads the pages out in front of you. You leaf through the six documents. Five
of the selected funds are in-house products of the advisor's bank, only one is
from an external fund manager.
153
e. High proactiveness, biased product mix, low persuasiveness
selection."
recommended.
f. Low proactiveness, balanced product mix, high persuasiveness
variety of fund profiles. He turns around to you and says: "Our in-house funds
154
are really excellent, I'll pick a few for you." You respond: "Yeah, sure,…but,
would it be possible that you also show me a few funds from other companies?"
Mr. Breiter hesitates briefly and answers: "Ok...sure…if you like…then I can do
that."
g. Low proactiveness, balanced product mix, low persuasiveness
that."
155
recommended.
h. Low proactiveness, biased product mix, high persuasiveness
that."
156
i. Low proactiveness, biased product mix, low persuasiveness
that."
recommended.
157
7.2 Scripts of the Video Treatments in Experiment 2
a. Introduction Text
Mr. Staiger has decided to put some money aside by taking out a life insurance.
It's important to him to choose the insurance product most appropriate for his
needs. Therefore, he's made an appointment with the local agency of a large
insurance company. A while ago, Mr. Staiger has read that some insurance firms
offer a so-called "open product architecture". Their customers can buy the firm's
own-label insurance policies, but also those of third-party providers. He intends
to find out whether his insurance company will also offer such third-party
products.
After Mr. Müller, the insurance agent, has welcomed him at his office, the two
of them discuss Mr. Staiger's financial requirements and expectations for a
while. Then, Mr. Müller suggests to present his customer with a selection of
insurance products that will meet his requirements.
b. Introduction Scene
Note: A (Agent), C (Client)
Client meeting room. Agent and client sit at a conference desk, facing each other
at a 90 degrees angle. The agent is nodding.
A:
"Allright. What I'll do now is to present a few different insurance options
that should be suitable for you."
C:
"Right, thanks."
158
c. Film with high proactiveness, balanced product mix, high persuasiveness
Scene 1: High Proactiveness
Fade-in.
Agent browses through a variety of brochures that are laid out on a sideboard.
He turns around to face the client.
A:
"You may have heard that our company offers an "open product
architecture"?
B:
"Yes, I've read that."
A:
"That means you don't get only in-house policies from us, but also those
of other providers. I'd suggest that I'll also include those third-party
products in my selection."
C:
"Yeah, I'd like that."
Fade-out.
Scene 2: Balanced Product Mix.
Fade-in.
With a focused expression, the client advisor's glance sweeps across the
different brochures. He picks one, puts it back. He picks another one, nodding.
Cut.
The agent takes the selected documents and walks over to the conference table,
where he spreads out 6 different brochures in front of the client.
A:
"Ok, let's have a look at this."
C:
"And these are all life-insurance products?"
A:
"That's right. What I've done is, I've picked three of our own insurance
products and another three of external providers, such as Allianz."
C:
"Okay, I see."
Fade-out.
159
Scene 3: High persuasiveness.
Fade-in.
The client sips at his water and takes a brief glance at the brochures
C:
"Ok…"
Client reclines in his chair.
C:
"Why do you recommend these particular policies? Could you explain
that in a bit more detail?"
The agent smiles and makes an inviting gesture.
A:
"Sure. I'll quickly give you an overview of the fundamental differences,
and then we can go into the detail, okay?"
C:
"Yeah, that's good."
The agent points at the individual brochures and elaborates in few words.
A:
"These three products here are classical endowment policies, in which the
payout sum depends almost exclusively on your fees. They vary in their
terms and conditions, though, so we'll have to look at this closely. And
these ones here are unit-linked life insurance policies. The return can be
much better with these, but also much worse, since part of their capital is
invested in mutual funds.
So much for a first overview – I'd suggest to go into the details now?"
C:
Fade-out.
d. Film with high proactiveness, balanced product mix, low persuasiveness
Fade-in.
A:
architecture"?
160
B:
A:
C:
Fade-out.
Fade-in.
Cut.
A:
C:
A:
C:
"Okay, I see."
Fade-out.
Scene 3: Low Persuasiveness of Reasoning
Fade-in.
C:
"Ok…"
C:
161
A:
"Well…"
Agent shrugs.
A:
"It's kind of difficult to explain that in detail."
Agent rubs his chin.
A:
"You know, there's a thousand different factors that play into this, right.
Like, should this be a mixed life insurance, or not. Is it date-fixed, and if
so, when? And should the whole thing feature investment funds? In that
case, you can later always make a shift from your assets…into the fund, I
mean. Or switching, there you make a reselection now and then."
C:
"Can you explain this in more detail?"
A:
"Well…they all have their own advantages. But you can trust me on this
one. I've got the right feeling for what will be the right thing for you."
Fade-out.
e. Film with high proactiveness, biased product mix, high persuasiveness
Fade-in.
A:
architecture"?
B:
A:
C:
Fade-out.
Scene 2: Biased Product Mix
162
Fade-in.
Cut.
A:
C:
A:
"That's right. What I've done is, I've picked five of our own insurance
products."
He points at one brochure.
A:
"And…then one more life insurance from another provider,…Allianz."
C:
"Okay, I see."
Fade-out.
Fade-in.
C:
"Ok…"
C:
A:
C:
163
A:
C:
Fade-out.
f. Film with high proactiveness, biased product mix, low persuasiveness
Fade-in.
A:
architecture"?
B:
A:
C:
Fade-out.
Fade-in.
164
Cut.
A:
C:
A:
products."
A:
C:
"Okay, I see."
Fade-out.
Fade-in.
C:
"Ok…"
C:
A:
"Well…"
Agent shrugs.
A:
A:
165
C:
A:
Fade-out.
g. Film with low proactiveness, balanced product mix, high persuasiveness
Scene 1: Low Proactiveness
Fade-in.
A:
"So, erm, the insurance products of our company are really excellent. I'll
pick a few for you right now."
C:
"Yeah, sure…but, would it be possible that you also present a few
products from other companies?"
The agent sighs and hesitates briefly.
A:
"Okay…yes…sure…if you like…I can do that."
C:
"Yes, please."
Fade-out.
Fade-in.
Cut.
A:
C:
166
A:
C:
"Okay, I see."
Fade-out.
Fade-in.
C:
"Ok…"
C:
A:
C:
A:
C:
Fade-out.
167
h. Film with low proactiveness, balanced product mix, low persuasiveness
Fade-in.
A:
C:
A:
C:
"Yes, please."
Fade-out.
Fade-in.
Cut.
A:
C:
A:
C:
"Okay, I see."
Fade-out.
168
Fade-in.
C:
"Ok…"
C:
A:
"Well…"
Agent shrugs.
A:
A:
C:
A:
Fade-out.
i. Film with low proactiveness, biased product mix, high persuasiveness
Fade-in.
A:
169
C:
A:
C:
"Yes, please."
Fade-out.
Fade-in.
Cut.
A:
C:
A:
products."
A:
C:
"Okay, I see."
Fade-out.
.
Fade-in.
C:
"Ok…"
170
C:
A:
C:
A:
C:
Fade-out.
j. Film with low proactiveness, biased product mix, low persuasiveness
Fade-in.
A:
C:
A:
C:
"Yes, please."
171
Fade-out.
Fade-in.
Cut.
A:
C:
A:
products."
A:
C:
"Okay, I see."
Fade-out.
.
Fade-in.
C:
"Ok…"
C:
A:
"Well…"
172
Agent shrugs.
A:
A:
C:
A:
Fade-out.
173
Curriculum Vitae
Name
Winfried Daun
Date of Birth
18 February 1975 in Solingen, Germany
Education
2008 - 2011
University of St. Gallen, Switzerland
Doctoral Candidate in Business Administration
1997
Lund University, School of Economics and Management,
Lund, Sweden
Courses in Cross-Cultural Marketing und HumanComputer Interaction
1994 - 2000
University of Passau, Germany
Diplom-Kaufmann
1985 - 1994
Gymnasium Schwertstrasse, Solingen, Germany
Abitur
Work Experience
2011 -
UBS AG, Zurich, Switzerland
Director, Head Marketing Strategy and Development
2007 - 2011
UBS AG, Zurich, Switzerland
Director, Senior Branding Spezialist
2004 - 2006
BBDO Consulting Suisse AG, Zurich, Switzerland
Manager
2000 - 2004
PricewaterhouseCoopers Unternehmensberatung GmbH
Consultant

The Effect of Salesperson Behavior and the Influence

Transcrição

Documentos relacionados

1pPPb22 - HASLab

2010 Colloquium Program - Ohio Northern University

business development - Plancorp Merchant Bank

INCT-MACC Annual Report

institutos nacionais de ciência e tecnologia – inct - INCT-Fcx

The Benefits of Sketching for Knowledge Management

PDNA - Centro de Conocimiento en Salud Pública y Desastres

NATIONAL SECURITY COUNCIL