Trading "best practices" - A good practice?

Abstract

In this article, we analyze approaches for identifying and transferring “best practices.” The relevant literature shows that
both, the identification and the transfer practices, are extremely difficult processes. Nevertheless, management consultants
position themselves and are seen by others as specialists in identifying and transferring effective or even best practices.
We analyze qualitative (interviews) and quantitative (survey) data on the ways in which German management consultants specializing
on strategy and process design try to identify and transfer “best practices.” Our results document that consultants neither
agree on the methods with which to identify best practices nor on the possibilities to transfer them between organizations
within one or between different industries. We show that attempts to transfer best practices are hampered with severe stickiness—even
for consultants. We conclude that best practices are mere marketing constructs of management consultants and suggest reframing
discussions on them around organizational routines and rules.

Full text

Industrial and Corporate Change, Volume 20, Number 3, pp. 683–719
doi:10.1093/icc/dtr011
Advance Access published April 9, 2011
Trading “best practices”—a good
practice?
Benjamin Wellstein* and Alfred Kieser**
,
y
In this article, we analyze approaches for identifying and transferring “best prac-
tices.” The relevant literature shows that both, the identification and the transfer
practices, are extremely difficult processes. Nevertheless, management consultants
position themselves and are seen by others as specialists in identifying and trans-
ferring effective or even best practices. We analyze qualitative (interviews) and
quantitative (survey) data on the ways in which German management consultants
specializing on strategy and process design try to identify and transfer “best prac-
tices.” Our results document that consultants neither agree on the methods with
which to identify best practices nor on the possibilities to transfer them between
organizations within one or between different industries. We show that attempts
to transfer best practices are hampered with severe stickiness—even for consult-
ants. We conclude that best practices are mere marketing constructs of manage-
ment consultants and suggest reframing discussions on them around
organizational routines and rules.
JEL classification: M10.
1. Introduction: has anyone ever seen a best practice?
Organizations use practices. They do the same things in different ways. For example,
they apply a variety of practices for selecting applicants, handling customer com-
plaints, choosing between investment projects, developing new products, conceptua-
lizing strategies, or organizing their production function. The performance of each
practice and the specific mixture of practices determines organizational
*Benjamin Wellstein, Institute for Business to Business I4B2B GmbH, PO Box 10 02, D-56171
Vallendar, Germany. e-mail: benjamin.wellstein@i4b2b.com
**Alfred Kieser, Faculty of Business Administration, University of Mannheim, D-68131 Mannheim,
Germany. e-mail: kieser@bwl.uni-mannheim.de
y
Main author for correspondence.
ß The Author 2011. Published by Oxford University Press on behalf of Associazione ICC. All rights reserved.
at University Mannheim on March 27, 2015http://icc.oxfordjournals.org/Downloaded from

competitiveness (Ray et al., 2004). Successful organizations are supposed to dispose
over at least a critical mass of superior practices.
Organizational routines defined by Winter (1964: 263) as “pattern of behavior
that is followed repeatedly, but is subject to change if conditions change” are basic
elements of practices. Frequently, organizations issue rules or standard operating
procedures in order to establish a practice (Cyert and March, 1963; March et al.,
2000). However, the information contained in rules in most cases specifies activities
only incompletely (Kieser and Koch, 2008) so that organizational members have to
develop routines when turning rules into action. Organizational members who are
socialized into a routine mostly (not always) also learn in which way the routine
represents “successful solutions to particular problems” (Dosi et al., 1992: 191–192).
In this way, routines are repositories of—tacit and explicit—organizational know-
ledge (Nelson and Winter, 1982; March, 1994; Miner, 1994; Hodgson, 1998; March
et al., 2000; Zollo and Winter, 2002).
Organizations strive to implement as many “good,” if not “best” practices as
possible in order to maximize their competitive advantage. A “best practice” has
been defined as “a set of interrelated work activities repeatedly utilized by individuals
or groups that a body of knowledge demonstrates will yield an optimal result”
(Tucker et al., 2007: 894). Unfortunately, not only the identification of “good” or
“best” practices but also their transfer between organizations confronts organizations
with severe problems. Astonishingly, in the academic literature, the problem of
transferring practices has received a lot more attention than that of identifying
good or best practices.
Conceptions of practices vary with regard to detail of the description. Rough
conceptions are used, for example, when corporate social responsibility (Wailes
and Michelson, 2008) or the shareholder value concept (Matzler et al., 2005) are
sweepingly denoted as best practices. In contrast, detailed descriptions of processes in
medical treatment (Tucker et al., 2007) are examples of more in-depth conceptions
of practices.
In the literature, some spectacular examples of the identification and transfer of
best practices can be found, lean production probably being the most famous one
(Womack et al., 1990). A team of the Sloan School of Management at MIT con-
densed the practices of Japanese producers of automobiles and recommended auto-
mobile producers in other countries to adopt these principles. The major automotive
companies (some of whom had financed the study) readily followed this plea. The
MIT concept became the world standard for automobile production (Krafcik, 1988).
The principles have also been transferred to other industries (Billesbach, 1994;
Dimancescu et al., 1997).
However, it is not at all clear whether this case is an example of a successful
transfer of a practice. The MIT team has been accused of constructing a highly
stylized model of the Japanese automotive production system there by downplaying
the specific Japanese context of automobile production (Sewell and Wikinson, 1992;
684 B. Wellstein and A. Kieser
at University Mannheim on March 27, 2015http://icc.oxfordjournals.org/Downloaded from

Williams et al., 1992; Rinehart et al., 1997). Automotive companies all over the world
that intended to import this concept had to modify and specify it to get it imple-
mented in their specific contexts (Scarbrough and Terry, 1998; Liker et al., 1999;
Woywode, 2002). There is no consensus whether and how lean production positively
affects performance (Williams et al., 1992). Lean production and other “best prac-
tices” of that sort are rather raw sketches of good rather than actually best practices.
The article is structured in nine parts. First, we discuss different approaches for
identifying best practices: expert judgment based on empirical evidence, success
factor research, and benchmarking. Then we deal with the problem of transferring
practices. There upon we show that consultants claim to be able to identify best
practices as well as to transfer them between organizations belonging to the same or
different industries—a maneuver we assume to be intricate. Synthesizing our con-
ceptual discussion, we develop four propositions on the identification, definition,
transferability, and stickiness of best practices by management consultants that guide
our empirical studies. Subsequently, we describe the samples and methods for our
qualitative and quantitative studies. On the basis of our empirical analysis, we de-
velop answers to our research questions and discuss the findings in a more general
way. We conclude with limitations of our study and implications for practitioners
and management research.
2. On the identification of good and best practices
Based on a literature review three main types of approaches for identifying good or
best practices can be differentiated: (i) expert judgment based on empirical evidence;
(ii) success factor research; and (iii) benchmarking.
2.1 Identifying good practices though expert judgment based on empirical
evidence
This method is common in medicine. Medical good or best practices are usually
described in great detail and are made available to everybody in this field who is
involved in the respective treatment. The usual process is that experts appointed by
medical associations screen and evaluate available empirical evidence and establish
consensus on good or best practices and recommend them to hospitals or practicing
physicians (see e.g. American Diabetes Association, 2005). That hospitals and practi-
cing physicians are not competing against each other facilitates sharing this know-
ledge. There is a general interest across the field to standardize and spread good
practices of treatment and patient care. However, due to differences in the evaluation
of the methodological quality of available studies, experts and therefore also medical
guidelines often disagree even “when the same randomized controlled trial (RCT)
evidence is cited” (McAlister et al., 2007: 1; see also Shaneyfelt et al., 1999). The
example of medical best practices shows how difficult it is to identify best versions
Trading “best practices”—a good practice? 685
at University Mannheim on March 27, 2015http://icc.oxfordjournals.org/Downloaded from

even for practices that are rather narrowly defined and described in great detail and
even if a large number of organizations, supported by scientific institutes and med-
iating associations, share information on the performance of alternative practices.
2.2 Identifying good practices through success factor research
Management scholars feel committed to fulfilling the “implicit promise of the code
(of the AoM to improve the effectiveness of organizations through our teaching,
research, and practice of management), the hopes of managers, the logic of perform-
ance improvement, and the ambitions of students of organizational adaptation”
(March and Sutton, 1997: 699). Success factor research is one of their preferred
approaches to live up to these expectations. This approach is applied to all organ-
izational functions, for example, to strategy (Meyer, 1991; Hambrick, 2007), HR
(Huselid et al., 1997) or manufacturing (Flynn et al., 1995). However, the expect-
ation that success factor studies are able to isolate best practices has been deeply
disappointed. In spite of the worldwide engagement of thousands of management
scholars in success factor research, consensus on any success factor or best practice
has not been reached yet. For example, since the 1970s, substantial research has been
invested in answering the question whether the practice of using formal strategic
planning procedures is advantageous—is a best practice—or not (Mintzberg, 1994).
The results are contradictory. A stable correlation between formal planning efforts
and company performance has not yet been established.
Research aiming at identifying best practices inevitably produces contradictory
results (Lampel and Shapira, 1995). Early research is criticized for its simplistic
methodology. Studies based on more sophisticated concepts and methodologies
come up with results that contradict earlier analyses. The original model is extended
by introducing more and more contingencies (Lampel and Shapira, 1995: 128).
Attempts to achieve convergence with the help of meta-analyses are frustrating be-
cause a generally approved methodology of meta-analysis does not exist (Bijmolt and
Pieters, 2001). Thus, the body of knowledge on how to solve specific problems is
getting more and more complex making it impossible to establish a relationship
between the contingencies identified in diverse academic studies and the specific
conditions that prevail in a company looking for a superior practice (March and
Sutton, 1997).
While some scholars argue that, over time, with the development of new methods,
consensus of results on best or, at least, better practices will be achievable (Rousseau
et al., 2008), other scholars maintain that the idea of identifying best practices on the
basis of empirical studies is fundamentally mistaken. Practices are unique, they
argue, and their impact on performance depends in complicated ways on their
interdependence with other practices (see, e.g. Starbuck, 1993).
Also, the resource-based view rests on the assumption that only practices
that other companies cannot easily reproduce competitive advantages (Dess and
686 B. Wellstein and A. Kieser
at University Mannheim on March 27, 2015http://icc.oxfordjournals.org/Downloaded from

Davis, 1984; Barney, 1991). Performance of some practices often affects success
of other practices (Denrell, 2005). Another point of critique is that best practices
tend to destroy themselves (March and Sutton, 1997). Practices that most
companies use no longer provide competitive advantages. Moreover, best prac-
tice studies can only analyze historical data. One cannot assume that practices
that worked well in the past will do so in the future (March und Sutton, 1997:
700). In addition, it is not possible to systematically include the risk that certain
practices entail for organizations in empirical analyses (Denrell, 2005). Another
problem is that performance criteria applied in a study are not necessarily the
performance criteria of a company interested in importing “best practices”
(Meyer, 1994).
We conclude that descriptions of practices in empirical success factor studies are
far too coarse. For example, when researchers analyze the impact of the effect of
diversification on organizational performance (Palich et al., 2000) they usually meas-
ure diversification with indices based on the product portfolio (Sambharya, 2000)
thereby abstracting from the routines that bring about, adjust, and maintain diver-
sification in organizations. The causal mechanisms of success or failure of diversifi-
cation are not disclosed (March and Sutton, 1997). The same kind and degree of
diversification can be managed more or less efficiently (if diversification across or-
ganizations is at all comparable). Therefore a correlation between the degree of
diversification and organizational performance does not say much about causalities
nor about rules and routines that successfully manage diversification. No wonder
that more and more management scholars doubt the usefulness of success factor
research (Camerer and Fahey, 1988; Daft and Buenger, 1990; Ghemawat, 1991; Kieser
and Nicolai, 2005).
2.3 Identifying good practices through benchmarking
Management consultants’ search for presumably best practices is not burdened with
the criterion of scientific rigor. They strive to offer problem solutions clients are
prepared to pay for. Brokering practices is easier when there is a story to tell that a
superior practice has been found systematically. Benchmarking provides such a story.
A typical benchmarking study measures performance of organizations or organiza-
tional units, and tries to identify the specific practices that are responsible for per-
formance differences (Camp, 1995; Bendell et al., 1998). The practices of the best
performers that explain the deviation to the second best performers are the best
practices.
However, this approach is problematic in several respects: it remains questionable
whether the chosen benchmark sample includes practices of superior performance.
And the practices of the benchmark participants have to be comparable so that they
can be evaluated along the same criteria that are relevant for the organization that is
interested in adopting the best practice (Holloway et al., 1999; Wareham and Gerrits,
Trading “best practices”—a good practice? 687
at University Mannheim on March 27, 2015http://icc.oxfordjournals.org/Downloaded from

1999; Bretschneider et al., 2005; Dani et al., 2006; Francis and Holloway, 2007).
And there is the problem that practices are context dependent (Davies and
Kochhar, 2002), that their influence on performance is mediated by other practices
they are interacting with.
The model that contains the key performance indicators and also assumptions
concerning the ways in which process design impacts these indicators cannot be very
complex since complexity reduces its chances of acceptance (Kaplan, 2003). As the
samples of benchmark studies are usually small, this model cannot be empirically
tested. Plausibility in the eyes of the managers involved in the study remains the only
validity criterion. In the integration phase, practices as assessed in the benchmark
company usually are modified in order to adapt them to the conditions of the
learning organization.
It is obvious that, seen from the perspective of management science, this approach
is intuitive rather than analytic. However, it has to appear rational in order to be
accepted by the managers who commissioned the benchmark project or the members
who are affected by it. Walgenbach and Hegele (2001) surmise that the underlying
purpose of benchmarking is not the improvement of practices through studying
better practices but establishing higher performance standards and thus placing
pressure on employees to improve the processes under study, regardless of whether
learning from benchmark companies is possible.
3. On the transfer of practices
As studies of learning new practices in hospitals show (Edmondson et al., 2001;
Berta and Baker, 2004; Tucker et al., 2007), organizational teams encounter
great difficulties in implementing new practices even if they are described in great
detail and with a high degree of accuracy. Existing routines and status relation-
ships in a team often present a powerful barriers in implementing new prac-
tices that can only be overcome with the help of specific training (Edmondson
et al., 2001). Learning new practices is also impeded by the fact that knowledge
about a practice and the practice itself are inextricably connected: “a view of
knowing as enacted in practice does not view competence as something that is
to be ‘transferred,’ and suggests that the very notion of ‘best practices’ is problem-
atic” (Orlikowski, 2002: 253). Also, routines and their context are inseparable: “when
removed from their original context, routines may be largely meaningless” (Becker,
2004).
Szulanski and his colleagues (Szulanski and Cappetta, 2003; Jensen and Szulanski,
2007; Szulanski and Jensen, 2008) have analyzed the problems of transferring prac-
tices under relatively advantageous conditions: between units with some similarities
in their contexts and with easy access to information, for example, between subsi-
diaries of multinational organizations or in franchising networks. As possible
688 B. Wellstein and A. Kieser
at University Mannheim on March 27, 2015http://icc.oxfordjournals.org/Downloaded from

impediments for such a transfer they identified a number of factors like causal am-
biguity, barren organizational context, arduous relationships, lack of motivation on
the side of the source or the receiver, or insufficient perceived trustworthiness of the
source. In different studies they found support for the existence of these factors
(Szulanski and Cappetta, 2003; Szulanski and Jensen, 2008). They also point out
that access to templates—working examples of the organizational practices—effect-
ively “helps overcome resistance by demonstrating results and providing evidence of
efficacy” (Jensen and Szulanski, 2007: 1727). Another result of their research is that
adaptation to the context of the receiving unit might decrease probability of transfer
success (Jensen and Szulanski, 2004).
In similar vein, Morris et al. (2009) analyze replication capability and common
practice use of HR practices within multinational companies. In their model, repli-
cation capability and the common use of a practice depends on three factors: people
alignment, that is shared understanding or vision, application process alignment,
that is the common use of guidelines, and information system alignment, that is
subsidiaries’ unified use of company-wide integrated information systems. Results
suggest that lacking alignment of people, processes, and systems can reduce replica-
tion capability and therefore the common use of a practice.
The transfer of practices between independent organizations, however, can be
supposed to be much more problematic than the transfer of practices between
highly comparable units like hospitals, subsidiaries of MNCs or the hub and the
nods of franchise systems. Nevertheless, there are, as we pointed out above, some
seemingly striking examples of a transfer of practices between organizations in highly
different contexts with lean production as perhaps the most famous of them
(Womack et al., 1990).
4. Consultants as brokers of best practices
Consultancies advertise possessing knowledge about good or best practices and being
able to let clients profit from it. For example, Boston Consulting Group, on its
Chinese subsidiary website, promises (Boston Consulting Group, 2009): “We help
them [our clients] design business strategies by leveraging our international best
practices and on-the-ground experience and know how.” One of Arthur D. Little’s
Managing Directors in the Middle East proclaimed in his inauguration speech in
2007 in Dubai that: “the international expertise of Arthur D. Little’s consultants from
South Asia, Europe and the United States ...enables us to help companies in the
Middle East improve their competitiveness by learning from best practices employed
by their industry peers across the world” (Arthur D. Little, 2009). In the McKinsey
Quarterly one can read that best practice “may be the most readily recognized and
widely used of all business management tools. ...To executives, modeling a com-
pany’s performance on its best-in-class competitor is an ambitious but attainable
Trading “best practices”—a good practice? 689
at University Mannheim on March 27, 2015http://icc.oxfordjournals.org/Downloaded from

aspiration” (Nattermann, 2000). Accenture has published a “Global Chemical
Industry Supply Chain Best Practices Study” (Accenture, 2008: 44). A. T. Kearney
has also published on best practices (Blascovich, 2005).
At first sight, the logic of best practice as propagated by consultancies seems
irresistible: by copying practices of the worldwide most successful companies, draw-
ing on consultants’ experience, one can improve one’s performance. However, some
authors warn that the practice of implementing best practices is not as easy as con-
sultants insinuate. Thus, Sanwal (2008) argues that “[o]ne of the most pervasive and
damaging follower afflictions that has increasingly infested corporate psychology is a
disease known as best practicism. ” He attributes the causes of this “disease” to a
number of erroneous beliefs, among them the belief that “best practices are easier
and less risky” (Sanwal, 2008: 52) and the belief that best practices can easily be
transferred from one company to another. In a similar vein, Wareham and Gerrits
(1999) question whether best practices remain best practices when transferred to
another context: “[A] bank pursuing a low cost strategy might think that the cost
of ‘selling a loan’ is an important yardstick. This measure will not be applicable in a
bank that is pursuing delivery of high-quality service for affluent clients who appre-
ciate the quality above the low cost of a service” (Wareham and Gerrits, 1999). Even
if the goals are comparable, they argue, embedding practices into another organiza-
tion is likely to reduce their effectiveness.
5. Research propositions
On the bottom line of your research we assume that consultants, in their attempts to
identify best practices and to transfer them to clients, are confronted with insur-
mountable difficulties so that clients cannot rely on improving performance as a
result of these endeavors. Considering the problems in identifying best practices and
in transferring them it is somewhat astonishing that best practices are given such a
prominent role in consultancies’ communication. Have consultancies found the al-
chemy to track best practices and to transfer them to clients? With our study, we
intend to shed some light on consultancies’ practices with regard to their attempts to
identify and transfer best practices. Therefore we synthesize our conceptual work into
the following propositions to guide our analyses:
P1: Consultants prefer benchmarking as an approach for identifying best
practices but also experiment with other approaches.
Benchmarking is an approach that many consultants sell to their clients. We there-
fore assume a preference for this approach in their efforts to identify best practices. If
clients or groups of clients do not ask consultancies to run benchmarking studies for
them, consultants can, in principle, apply benchmarking to practices they come
across in their projects with different clients.
690 B. Wellstein and A. Kieser
at University Mannheim on March 27, 2015http://icc.oxfordjournals.org/Downloaded from

At the same time, we expect that consultants are aware of the shortcomings of
this approach and might therefore experiment with alternative approaches. They
might, for instance, use data collected in client companies for some kind of success
factor studies. As we have pointed out, they publish studies of this sort or they
conceptualize best practices as a rough concept like lean production: a concept
that results from the consultants’ stylized interpretation of practices of successful
clients.
P2: Consultants hold different definitions of best practices.
We assume that consultants, in principle, in their attempts to conceptualize best
practices, are facing difficulties similar to those of success factor researchers.
However, how consultants define practices influences their framing of client prob-
lems and the ways they go about solving them. Hence, we postulate that there is no
consensus among consultants on the definition of best practices and how to build up
a knowledge basis by analyzing the influence of best practices on organizational
performance.
P3: Consultants hold different views on the transferability of practices be-
tween organizations within one or between different industries.
Our discussion has shown that even transferring practices within one company
respectively very similar organizations (e.g. hospitals) is highly problematic.
Difficulties can be assumed to maintain or even increase when transferring
practices with the help of a knowledge broker between different organizations
of the same industry or between organizations of different industries. Consultants
imply to be able of resolving transfers of the last two categories. We want to
find out to what extent consultants are aware of difficulties in comparing and
transferring practices across different companies within one or between different
industries.
P4: Consultants tend to interpret best practices as solution patterns that can
be captured in abstract representations serving as inspiration for the solution
of similar problems in other organizations—an operation we postulate to be
sticky.
Consulting companies possess internal structures and systems to leverage the know-
ledge on practices they come across in benchmarking exercises or project work in the
field. Big consulting companies often establish practice groups (Anand et al., 2007).
Visible for the client, however, is just the knowledge a consulting company brings
into a project. This knowledge stems from personal experience of individual con-
sultants or project teams and is transferred within a consultancy by the means of
communication or documentation on solution patterns. In order to inspire their
colleagues who might get confronted to similar problems, practices are described in a
Trading “best practices”—a good practice? 691
at University Mannheim on March 27, 2015http://icc.oxfordjournals.org/Downloaded from

general way that abstracts from prevailing conditions in client organizations and
stored in an information system.
As the transfer of practices within organizations has been shown to be sticky, we
assume that this is also true for consultants’ knowledge about “best practices.” Each
implementation of an abstract concept is a risky endeavor requiring the design of
new task structures, the training of employees, their development of new routines.
Since unintended effects can never be fully avoided, observations of deviating results
necessitating corrections are likely to occur. We therefore assume that the problems
that cause stickiness of intraorganizational transfers are inherent to a higher degree in
interorganizational transfers of practices. For our study of consultants’ practices of
practice transfer we thus apply findings from the research on sticky knowledge by
Szulanski et al. (Szulanski and Cappetta, 2003; Jensen and Szulanski, 2007; Szulanski
and Jensen, 2008).
Figure 1 shows the framework of our analyses entailing two primary dimensions.
The first dimension “origins of stickiness” in principle draws on the
knowledge-based view (e.g. Mahoney and Pandian, 1992; Decarolis and Deeds,
1999) in which ambiguity (Lippmann and Rumelt, 1982; Reed and DeFillippi,
1990; Simonin, 1999) and absorptive capacity (Cohen and Levinthal, 1990: 128)
are emphasized as important factors influencing knowledge transfer. In addition
we include a factor that did not have a major impact in intraorganizational transfers:
“unproven knowledge.” As Szulanski (1996) coins it: “[K]nowledge with a proven
record of past usefulness is less difficult to transfer.” In the field of management
consulting this applies in particular as consultants like to ensure their knowledge on
practices is of superior quality.
With the second dimension, we analyze the actual stickiness of practices circulat-
ing within a consulting company. This measurement is based on the insight that
practices are stripped of their original context and communicated in such a way that
others can apply and adjust them to a new context. This is a highly problematic
attempt and thus we assume that consultancies are facing problems in documenting
and communicating practices to such a detail that replication in other contexts is
possible. We also include context specify as impediment of such replication (“barren
context”). The similarities and distances to Szulanski et al.’s studies (Szulanski and
Cappetta, 2003; Jensen and Szulanski, 2007; Szulanski and Jensen, 2008) on stickiness
Figure 1 Framework for assessing the stickiness of best practices (transfer).
692 B. Wellstein and A. Kieser
at University Mannheim on March 27, 2015http://icc.oxfordjournals.org/Downloaded from

of the transfer of best practices within one organization we will explain in more detail
in Section 7.2, when we describe the development and validation of our measure-
ment scales.
Overall we hypothesize positive links between the constructs that we identified
and their higher dimensions, for example higher causal ambiguity contributing to
more of the origins of stickiness leading in turn to more stickiness reflected in the
stickiness of documented knowledge on best practices available within a consultancy.
The higher the degree of the origins of stickiness, the less we assume consultants
being able to find, understand and absorb best practices leading to a higher actual
stickiness of practices.
6. Sample and methods
In our research, we combined qualitative and quantitative studies (see e.g. King and
Zeithaml, 2003) using data from samples of German management consultants. Such
a “triangulation” (Denzin, 1978) enables researchers to base quantitative represen-
tative analyses on insights gained through in-depth observations in the field. Thus, in
interviews with consultants, we reconstructed how the interviewees perceived oper-
ations around best practices. On the basis of the results of this study, we developed a
quantitative analysis aiming at a representative overview over the ways in which
consultants attempt to identify and transfer best practices.
6.1 Qualitative study
Between March and September 2006, we conducted semi-structured interviews with
12 consultants and 2 knowledge managers from different consultancies. Table 1 lists
the position and the interviewees’ extent of professional experience (sorted by the
firm sizes of the consultancies). Our sample contains members of consultancies from
different hierarchical levels ranging from junior consultants up to senior consultants
and members of management boards. Six of the 10 consultants, we interviewed
worked in large, international consulting companies. Two interviewees worked in
medium-sized and two in small consultancies (up to 10 consultants). One of the two
knowledge managers we interviewed was employed by an international consulting
firm and the other one was working for an internal consulting agency of a global
company. The interviews lasted between 45 and 90 min each and were conducted in
person (five) or over telephone (seven). In two cases, the interviewees did not allow
recording. In these cases, we reconstructed the interviews on the basis of notes
generated during and after the interviews.
We used the ATLAS.ti 5.0 software for analyzing the transcripts of the interviews
following Miles and Huberman’s (1994) recommendations. At first we assigned
codes to text passages in order to classify them into categories of topics. In an
Trading “best practices”—a good practice? 693
at University Mannheim on March 27, 2015http://icc.oxfordjournals.org/Downloaded from

iterative manner, we screened the data, compared statements, summarized them and,
in a cyclical manner, arrived at conclusions. The statements quoted in this article are
the most representative ones supporting our explanations.
6.2 Quantitative study
On the basis of qualitative insights we developed a standardized questionnaire that
we pretested with four consultants using the “think-aloud” method (Ericsson and
Simon, 1980; Strack and Martin, 1987; Ericsson and Simon, 1993) leading to changes
in the wording of items. From several sources, we compiled a sample of 274 con-
sulting companies operating in the German market consulting on strategy or on
process and organizational design. The focus was on medium-sized (i.e. more than
seven consultants) up to bigger consulting companies including bigger in-house
consulting departments of German global companies. We identified the names of
1.130 consultants working for these companies. In order to reduce key informant
bias we tried to identify at least two informants in each company. Within bigger
consulting companies we included two or more consultants from different
Table 1 Sample composition—qualitative study
Interview partner(s) Position and company Consulting
experience
(years)
C1 and C2 Consultants of a globally leading consulting firm (top 5) Both 1.5
C3 Junior Project Manager of a globally leading consulting
firm (top 5)
3.25
C4 Junior Partner of a globally leading consulting firm (top 5) 6
C5 Partner of a large global consulting company 12
C6 Senior Project Manager of a large global consulting
company
8
C7 Senior Consultant of a large global consulting company 5
C8 and C9 Managing Director and Project Leader of a medium-sized
consultancy
15 and 5
C10 Managing Director of a medium-sized consultancy 8
C11 Owner of a small consultancy 13
C12 Owner of a small consultancy 21
K1 Knowledge Manager Central Europe of a large global
consulting company
5.25
K2 Knowledge Manager of the large internal consultancy of a
global player
5
694 B. Wellstein and A. Kieser
at University Mannheim on March 27, 2015http://icc.oxfordjournals.org/Downloaded from

hierarchical levels. We did not address knowledge managers; however, some inform-
ants who had received our questionnaire forwarded it to a member of their internal
knowledge management department to answer it.
Between August and October 2007, we sent out invitation letters by mail, followed
by emails that contained an electronic version of the questionnaire and a link for
answering the survey over the internet. During this phase, we dropped 7 consulting
companies and 21 consultants since they concentrated on HR consulting or had left
their companies, respectively. The final sample size was therefore 1.109 consultants
from 267 consulting firms.
We obtained 232 completed and 72 partly completed questionnaires (450% of the
questions answered) from 152 consulting companies. Since the questionnaire en-
tailed items of measures on knowledge management in addition to the measures we
concentrate on in this article, we included the partly completed questionnaires in our
analysis only if we found comprehensive answers in the section on best practices. The
response rate was 57% in terms of the consulting companies and 27% in terms of the
individual consultants surveyed (final sample sizes including partly completed ques-
tionnaires). All of the top 25 consulting companies in the German market are rep-
resented in our study. Prior to the analyses, we dropped nine answers from multiple
respondents (i.e. companies from we received more than one answer) and two sin-
gular answers (i.e. companies from which we received just one answer) as those
participants had categorized themselves as not being sufficiently confident with
their responses. Thus, our data come from 150 consulting companies. We asked
whether participants frequently dealt with best practices in their project work, and
157 participants out of 96 companies stated that they did. We selected these 157
consultants for our study (Table 2). Our study has been successful to the extent that
it reports on large to medium-sized consulting companies that primarily offer con-
sulting on strategy or on process and organizational design. Most of our informants
occupy the hierarchical position of partner. On average, our respondents are looking
back on 7 years of consulting experience.
We ran several nonresponse analyses (t-tests, linear regression) and found
no systematic significant deviation between early and late responses (Armstrong
and Overton, 1977; Dooley and Lindner, 2003; Rogelberg et al., 2003). As we
had numerous responses from respondents working in the same company, we
checked for consistency using the item means-based average deviation index
(Burke and Dunlap, 2002). Significant disagreement between the raters led to elim-
ination of the respective answers. Otherwise we used the mean values of the raters’
answers.
The analyses we deployed extended from simple frequency scales to exploratory
factor analyses and canonical correlation analyses (e.g. Thorndike, 2000; Sherry and
Henson, 2005). We will go into more detail on our procedures when we present our
findings.
Trading “best practices”—a good practice? 695
at University Mannheim on March 27, 2015http://icc.oxfordjournals.org/Downloaded from

Table 2 Breakdown of responses to the quantitative study
Firm size (96) Service lines (96) Position (157)
Large Over 500 consultants 5 Strategy-, process-, and organization consulting 52 Partners 42
500–101 consultants 14 Process- and organization consulting 26
100–51 consultants 25 Strategy consulting 5 Engagement Managers 26
Mid-size 50–31 consultants 20 IT-, process-, and organization consulting 7
30–21 consultants 14 IT-, strategy-, process-, and organization consulting 5 Consultants 30
20–11 consultants 13
10–6 consultants 6 HR-, process-, and organization consulting 2 Knowledge Managers 2
Small Up to 5 consultants 3 Other combinations 3
All values in percentages of the top figures in parentheses.
696 B. Wellstein and A. Kieser
at University Mannheim on March 27, 2015http://icc.oxfordjournals.org/Downloaded from

7. Analyses and results
7.1 Consultants identifying and defining best practices
7.1.1 P1—identification of best practices
We could identify three groups that apply different approaches for identifying best
practices (see Table 3). The first group (“positivistic benchmarking”) identifies best
practices through benchmarking studies that we already categorized as consultants’
favorite approach. Interestingly, the conversation between Consultant 1 (C1) and
Consultant 2 (C2) in Table 3 reveals that consultants are may also be in conflict with
the claim that this approach is capable of identifying best practices.
A second group (“critical benchmarking”) expresses a somewhat critical view
toward benchmarking. First, they question whether a best solution exists at all.
Second, they admit that comparing solutions across industries and companies is
most problematic and, in most cases, not defendable. Third, they maintain that
there is no need to engage in comparisons of solutions across companies and indus-
tries in order to come up with best practices.
A third group of our interlocutors (“neglecting benchmarking”) does not mention
any specific method for hedging the status of their best practices. They argue that,
over time, it is possible through repeated observations to assess the performance of a
practice. Benchmarking requires access to well performing benchmarking partners.
Medium-sized consulting companies may not have the client base needed to get
access to a sufficient number of different practices. In this vein, interview partner
C12 questions the ability of smaller consulting companies to arrive at absolute judg-
ments about best practices although he presumes that globally operating competitors
might be in a position to be successful in this respect.
All in all, there seems to be a correspondence between the client base of a con-
sulting company and the attitude toward benchmarking as an appropriate method
for identifying best practices. Larger companies are engaged in many projects for
many clients and therefore seem to be—or at least see themselves—in a better pos-
ition to carry out systematic comparisons of practices. Smaller consultancies, how-
ever, are rarely contracted by a larger share of organizations in a particular industry
and therefore their client base is too small to leverage their insights. However, this
assumption drawn from our interviews is not directly supported by our quantitative
study in which we asked whether consultants use benchmarks for identifying best
practices. We did not find significant differences between the answers of larger and
medium-sized consulting firms.
Interestingly, one of our interview partners from a top-10 consulting firm de-
scribes a systematic approach for delivering evidence of success of practices that
reminds one of the logic of academic success factors research:
C5: “[A]t the end best practices are reflected in the performance.So
(...) we somewhere find a dimension that is quantitatively measurable,
Trading “best practices”—a good practice? 697
at University Mannheim on March 27, 2015http://icc.oxfordjournals.org/Downloaded from

Table 3 Consultants’ approaches to identifying best practices
Approach Data from interviews Interview partner
Positivistic benchmarking C2: So best practice is a very general term you can use for anything. One can say: “I derive Best practices
from benchmarks.” In this sense benchmarks are also best practice, for example regarding expense
ratios. But you can also say very operationally, best practice is an organizational structure or a process. So
with best practice not matter what topic it is, it means, there is a competitor that has so far resolved a
topic optimally in comparison to its competitors.
C1/C2
C1: Or better—achieves a better solution at least.
C2: Yes (...) but best practice is the optimal solution.
C1: Yes, but it never exists.
C2: That exists in the current market. *Now it is getting philosophical.
If one has a best, a comparable group, yes, then best practice is the one that performs best according to
the criteria of comparison, yes.
C4
And the group whose practices one compares is then given by the client base?
That is just the “high” art of benchmarking, to identify the practice.
Best practice is, among comparable processes, so to speak, the one that best meets the goals. C10
So benchmarking is generally a comparative analysis, but best practice is precisely the comparison with the
best of a group (...) so if someone performs particularly well in a process or when a company is
particularly successful, is profitable, so that you are always measuring with the best and look (...) who is
the best, then compare and see what are the reasons for this.
K2
(continued)
698 B. Wellstein and A. Kieser
at University Mannheim on March 27, 2015http://icc.oxfordjournals.org/Downloaded from

Table 3 Continued
Approach Data from interviews Interview partner
Critical benchmarking How did you notice a practice is actually “best”? C3
One cannot ascertain that, the impression counts. Success often shows with considerable delay and one
does not know the whole world.
So, for our company, best practices are the best known methods or means to realize things. Can one
possess best practices or can there be a best practice; that leads to, I’d say, philosophical discussions of
benchmarking and of who is making something better or less well. Let’s say as a strategist, I have exactly
this discussion very often with my clients: does it make sense—within discussions of strategy formula-
tions—to compare oneself with one’s competitors. What often happens [is] that a consultant defines
that what he proposes is best practice (...). It is just, excuse me, best practice is such a consultant
bullshit-bingo word.
C6
To what extent are benchmarks carried out with which a client proofs to possess best figures on certain
practices?
C5
It is always the case that he [the client] either says: “I am the best” or he says: “I am quite different.” So in
this respect it’s always very difficult to proof this point.
So this external benchmarking, this is very often demanded, but, as I think, this only works for very specific
processes and has actually little relevance for our work (...). To be precise, we use this term relatively
rarely (...). For sure, a best practice is an outstanding problem solver, however one often lumps together
companies that cannot be lumped together.
C11
(continued)
Trading “best practices”—a good practice? 699
at University Mannheim on March 27, 2015http://icc.oxfordjournals.org/Downloaded from

Table 3 Continued
Approach Data from interviews Interview partner
Neglecting benchmarking Time has shown that these practices have been quite good. “Best” practices I do not want to say, and one
reuses those “best” practices again. So, I think, at the beginning one does not recognize that there is a
best practice, but time tells and shows that it certainly can be implemented and used again.
C7
Well best, in the sense of “absolute best,” which is what was originally meant, I would not dare to say,
since eventually it is always a relative comparison with what one has seen so far. I am simply too small, I
am not operating on all continents with my consultancy. Those kinds of companies can perhaps judge
absolutely. I do not have these possibilities. Hence, I always have to lower my assumptions and have to
say: “It is only a relative best practice that I identified here.” Relative refers basically to an intertemporal
comparison: “Such a good solution I have not been seeing up to now.”
C12
I hear it again and again, consultants permanently ask for best practices. Well, let us say these are
applications or maybe processes that have been proven to be successful in the field several times and
seem therefore to be to be susceptible to be grouped together since promising high success.
K1
So do applications and processes also include benchmarks?
Not necessarily. You need benchmarks to see where you stand with, for example the costs for your
overhead, but benchmarks as such, I would not subsume under best practices.
700 B. Wellstein and A. Kieser
at University Mannheim on March 27, 2015http://icc.oxfordjournals.org/Downloaded from

for example, turnover or profit margin, when we then say “innovation
management works well’ there must be a correlation [between this
characteristic] and outstanding (...) industry-specific margins and
results. And on top there is then a credibility test whether this fits or
not. So that it not only looks great and feels good but you eventually see
it in your results.” [Emphasis added]
We find little agreement between consultants on how to identify best practices.
They tend to vaguely define best practices as practices that have proven “outstand-
ing,” as practices that are producing good results, or as practices that are reliably
connected to successful companies. The differences in definitions become obvious
when, for instance, interviewee C3 defines best practices as methods that encompass
processes, organizational forms, and technologies whereas consultant C5 argues that
best practices are models of approaches that entail processes, quality requirements,
and organizational roles.
7.1.2 P2—definitions of best practices
Prompted by these divergent patterns in our interviewees’ responses we listed the
elements of best practices that had occurred in our interviews with experts (i.e.
approaches/methods, processes, forms of organizing, roles, and technologies) in
our survey questionnaire. Based on this list, we asked participants to indicate what
they usually associate best practices with when using this expression in their projects.
The result of the analysis is depicted in Figure 2.
As a result of the expert interviews we proposed that there is no conver-
gent understanding of best practices prevalent in consultancies. The results of
the quantitative study support this proposition. Altogether we found 21 different
combinations of elements of best practices without one being dominant. However,
there are differences between the answers of consultants from large and
medium-sized consultancies. The most frequent combination of best practice elem-
ents selected by consultants from large consulting agencies (first in Figure 2) includes
a wide array of factors. In contrast, consultants from medium-sized consulting
companies rarely see best practices as encompassing such a diverse bundle of
elements.
Interestingly, we found little convergence when comparing answers of informants
from the same consultancies. In 68% of 34 companies, informants from the same
companies differed with regard to the elements included into their notions of best
practices. Without convergence in the interpretation of best practices within indi-
vidual consulting companies, agreement on what a best practice is across the market
is improbable. Perhaps a workable agreement on the definition of best practices can
be established on the basis of organizational routines. We will come back to this idea
in the Section 8.
Trading “best practices”—a good practice? 701
at University Mannheim on March 27, 2015http://icc.oxfordjournals.org/Downloaded from

7.2 Transfer of best practices
Concerning stickiness we present findings on three questions: (i) can practices be
transferred? (ii) How can factors influencing stickiness of practices be measured; and
(iii) Is there stickiness in the transfer of best practices? We also test our basic
hypotheses concerning stickiness.
7.2.1 P3—transferability of practices
Analogous to the previous section there is fundamental disagreement among our
interviewees whether practices are transferable across industries. First of all, there is
no consensus on whether best practices can be described in an abstract way without
referring to specific industries. The following quotes from two of our experts reflect
the diversity of assumptions concerning this issue:
C6: (...) [A]nd if one sees the evolution (...) in different industries, *
you have identified a best practice in a certain industry and in another
industry one would say: ‘This is an old hat’ [meaning that this practice is
most common].”
C1: “So a best practice for me is classically a solution that works in an
industry very well and is then transferred to another industry. Example:
sales systems respectively the sales management in the pharmaceutical
sector is extremely well developed and could very well be transferred to
the chemical industry. There are sales representatives in the
6%
6%
7%
8%
9%
13%
15%
Roles, processes, forms of
organizing, technologies and
approaches/methods
Processes, forms of organizing
and approaches/methods
Roles, processes, forms of organizing
and approaches/methods
Processes, forms of organizing,
Technologies and approaches/methods
Processes and
approaches/methods
Processes, technologies and
approaches/methods
Roles, processes and
approaches/methods
36%Other combinations
Large Consultancies
(114 participants)
Mid-Size Consultancies
(39 participants)
8%
10%
15%
5%
8%
10%
3%
41%
Figure 2 Variation across definitions of “best practice”.
702 B. Wellstein and A. Kieser
at University Mannheim on March 27, 2015http://icc.oxfordjournals.org/Downloaded from

pharmaceutical sector who excel and people in all other companies could
do so, too. So best practices are
hardly specific. [Emphasis added]
Prompted by these controversial views, we used the quantitative study to get
assessments from a larger group of consultants. We used a semantic differential
for asking participants whether best practices can be transferred between industries
or are only valid within single industries. Moreover, we asked if best practices are
specific for a company or can be used generally. The frequency distributions along
these two dimensions are depicted in Figure 3. They suggest two conclusions regard-
ing mean values and variance of distribution. The mean values show that consultants
tend to see best practices as solutions that are transferable between industries as well
as between organizations what is in line with their public announcements. Two
response groups are dominant, jointly accounting for 55% of the combined re-
sponses. The first group assumes that best practices are fully or almost fully trans-
ferable across industries and organizations. This assessment is, however, contrasted
by that of a larger group which is more cautious regarding transferability of practices
across industries and organizations.
Usable
generally
Frequency of
combinations
(155 participants)
Organization-
specific
Industry-
specific
Transferable
between
industries
Ø 69.0
25
20
15
10
5
1
Ø 64.1
100
83
67
50
33
17
0
10083675033170
11 711
33 4
24
2
12 8
2
121
9
1
23 16
2
1
21
15
5
8
3
3
1
1
Figure 3 Assumptions on transferability of best practices.
Trading “best practices”—a good practice? 703
at University Mannheim on March 27, 2015http://icc.oxfordjournals.org/Downloaded from

Apparently, best practices are social constructions of consultants who try to use
them as a sales pitch.
7.2.2 P4—stickiness of best practices: scale development and validation
We anticipated that consultants would perceive problems in extracting best practices
from clients and generalizing this knowledge in order to transfer it to other consult-
ants of the same consultancy who then apply them to other companies. As we
argued, practices are sticky in the sense that major problems exist in identifying,
incorporating and understanding the ways in which they are linked to performance.
However, this knowledge is essential for successfully transferring practices.
Even though we considered the identification of “best practices” as most prob-
lematic, in the formulation of our items we stuck to this expression since consultants
use it in their daily language. We used multiple items with 7-point rating scales
(anchors “strongly disagree” and “strongly agree”, respectively, “very seldom” and
“very often”). A complete list of items and their statistics can be found in the
Appendix A. To validate our measurement, we used common methods like
Cronbach’s , item-to-total correlations, and exploratory factor analyses (e.g.
Churchill, 1979). Finally, we computed a linear transformation of all values (eight
minus item value) so that lower values on the scales corresponded with less stickiness
and more vice versa.
Within the dimension of the “origins of stickiness,” we used and adapted items
from Szulanski’s (Szulanski, 1996, Szulanski and Cappetta, 2003; Szulanski et al.,
2004) research. Exploratory factor analysis extracted only one factor per construct.
The first factor, “unproven knowledge,” measures the stickiness of practices with
regard to the extent of difficulties to identify practices as “best” or superior practices.
The item that asks for the usefulness of benchmarking negatively affects reliability
(item-to-total correlation 0.31, Cronbach’s  at 0.68). Looking at the diversity of
opinions concerning this issue this finding is not surprising. After its elimination we
reach satisfying reliability levels (e.g. Bagozzi and Yi, 1988). The second factor,
“causal ambiguity,” tries to measure to what extent consultants are able or unable
to explicate the causality of positive effects of practices on performance. Finally, “lack
of absorptive capacity” measures the ability or disability of consulting companies to
absorb effective practices by documenting practices and anchoring them within their
organization, steps that precede transferring them from one context into another.
To measure our second dimension, “stickiness,” we developed new items. We had
to take into account that the transfer of this knowledge between clients is brokered by
consulting organizations. Szulanski’s (1996, Szulanski and Cappetta, 2003) intraor-
ganizational measurement within this dimension is therefore not directly applicable
and we use stickiness in documents, communication and context as proxies.
Compared to the mentioned research on the stickiness of practices we herewith
focus on a different object vis-a
`
-vis the practices under study. Whereas the measure-
ment of the origins of stickiness is in line with the intraorganizational research on the
704 B. Wellstein and A. Kieser
at University Mannheim on March 27, 2015http://icc.oxfordjournals.org/Downloaded from

transfer of best practices, we concentrate in this dimension on project situations in
which consultants have the chance to acquire knowledge on practices. By choosing
these different foci we try to overcome the risk of producing tautologies by measur-
ing the same thing.
We asked consultants whether the documents available in their consultancy
enable readers to understand the described practices, their causality and the measures
necessary to implement them (“sticky documented best practices”). Furthermore we
asked whether discussions on best practices inside the consulting companies offered
sufficient background information to warrant a successful transfer (“sticky commu-
nication on best practices”).
In each factor, we had to eliminate an item that compromised the measurement
because it triggered an answer in a direction inverted to that of the other items so
that reverse scoring would not have been appropriate. Finally, we measured the
extent to which the context sensitivity of best practices precluded their transfer
(“barren context of best practices”). Here, we identified to what extent practices
have to be modified in the process of their implementation into other contexts. If
consultants express the view that modifications are necessary we can conclude that
the respective practices are context sensitive and therefore sticky.
Exploratory factor analysis of individual constructs extracted only single factors.
Validating this measure we had to eliminate one item that referred to the
cross-industrial aspect of best practices. This elimination corresponds to the con-
sultants’ differing views on this issue.
Overall we obtained acceptable reliability levels for all factors in our two dimen-
sions (Table 4). A prerequisite for the intended analysis is that multi-collinearity is
not substantial, that is factors do not strongly correlate with each other. We calcu-
lated variance inflation factors (VIF) for every factor vis-a
`
-vis all other factors ran-
ging from 1.13 to 1.94 and staying clearly below a threshold of 10 (e.g. Chatterjee and
Price, 1977: 182; Kennedy, 1998: 90). Thus, we did not find an indication for sub-
stantial multicollinearity.
Finally we checked for discriminant validity (Table 5) which means that we
analyzed the shared variance between a factor and its related indicators (i.e. the
average variance extracted—AVE). It should not be larger than the shared variance
between a factor and the other factors (Fornell and Larcker, 1981). Based on a
comparison of the roots of AVE with the bivariate correlation coefficients the cri-
terion holds for all factors. Hence, discriminant validity is sufficient.
7.2.3 P4—stickiness of best practices: canonical correlation
In order to identify patterns in the factor structures of both dimensions, we ran a
canonical correlation analysis (Table 6). In case of missing values, we excluded cases
list wise, leading to 73 complete cases as empirical bases for our analyses. The head-
ings “roots,” respectively, “root no.” refer to function(s) of the canonical correlation
(Thorndike, 2000; Sherry and Henson, 2005). For any root combination, we present
Trading “best practices”—a good practice? 705
at University Mannheim on March 27, 2015http://icc.oxfordjournals.org/Downloaded from

Table 4 Measurement information for individual factors
Dimension Factor Description Items N  KMO AVE CR
Origins of stickiness Unproven knowledge Ability to recognize the superiority of a practice 3 91 0.71 0.64 0.64 0.84
Causal ambiguity Ability to understand the causality of a practice 7 93 0.90 0.90 0.62 0.92
Lack of absorptive capacity Ability to absorb the practice 5 92 0.82 0.83 0.58 0.87
Stickiness Sticky documented BP Quality of available documented best practices 4 83 0.83 0.78 0.68 0.89
Sticky communication on BP Quality of communication with colleagues on best practices 4 83 0.90 0.83 0.77 0.93
Barren context of BP Magnitude with which the BP has to be adapted when applied 2 89 0.65 0.50 0.74 0.85
: Cronbach’s alpha; KMO: Kaiser–Meyer-Olkin-measure; AVE: average variance extracted; CR: composite reliability.
706 B. Wellstein and A. Kieser
at University Mannheim on March 27, 2015http://icc.oxfordjournals.org/Downloaded from

test statistics (, 
2
, df) and the level of significance (P). Explained variance across all
three functions is 1
13
¼59%. The whole model (all three functions combined) is
statistically highly significant (P50.0001). We conclude, therefore, that we have
identified a substantial link between the two measured dimensions of the origins
of stickiness and stickiness of best practices.
A look at the level of single functions (“root no.”) reveals that the first ordination
of the canonical correlation analysis accounts for R
XY
2
¼51% of the shared variance
between our two dimensions.
1
With R
XY
¼0.71 there is a strong positive link be-
tween the two dimensions. We will therefore concentrate in our interpretation on the
first canonical function as its explained variance of 51% accounts for much of the
variance of the total model (59%). Respectively, the statistical significance levels of
the second and third functions drop below the threshold of 1%.
An interpretation of the selected canonical function should take the canonical
weights into consideration (Table 7). Canonical weights are in their interpretation
Table 5 Measurement information on multicollinearity and discriminant validity
ffiffiffiffiffiffiffiffiffi
AVE
p
123456
1 Unproven knowledge 0.80
2 Causal ambiguity 0.45** 0.79
3 Lack of absorptive capacity 0.43** 0.61** 0.76
4 Sticky documented best practices 0.43** 0.45** 0.60** 0.83
5 Sticky communication on best practices 0.37** 0.60** 0.52** 0.49** 0.88
6 Barren context of best practices 0.34** 0.25* 0.26* 0.36** n.s. 0.86
**P50.01; *P50.05.
n.s., not significant; AVE, average variance extracted.
Table 6 Results of canonical correlation analyses
Roots Wilks 
2
df P Root no. R
XY
R
XY
2
1–3 0.41 61.33 3 3 ¼9 0.000 1 0.71 0.51
2–3 0.83 12.38 2 2 ¼4 0.015 2 0.36 0.13
3–3 0.96 3.07 1 1 ¼1 0.080 3 0.21 0.04
1
The sum of explained variance of all three canonical functions (68%) is bigger than the explained
variance of the model (59%), because the former are orthogonal (hierarchical process).
Trading “best practices”—a good practice? 707
at University Mannheim on March 27, 2015http://icc.oxfordjournals.org/Downloaded from

analogous to standardized -coefficients of a regression analysis, that is the larger a
weight’s absolute value the more important it is for the corresponding factor. The
weights of the first canonical function can be interpreted along the two dimensions.
First, within the dimension “origins of stickiness,” we find “lack of absorptive
capacity,” that is the extent to which practices are absorbable by the consulting
companies, the greatest obstacle for transferring practices. There is also a strong
inhibiting effect due to “causal ambiguity,” that is the extent to which consultants
can fathom the causality structure of a practice that they identified as a best one. The
factor “unproven knowledge” is of minor importance in terms of the origins of
stickiness. This supports our claim that consultants seldom try to prove that their
practices are “best practices”—an operation that would be futile anyway. The second
dimension “stickiness” is explained very clearly by the difficulties of decoding and
implanting best practice knowledge. Context sensitivity, however, provides no sig-
nificant explanation within the model.
Results of the canonical correlation are summarized in Figure 4 (values from the
first ordination of the canonical correlation). To conclude, there is a strong and
statistically significant positive interrelation between the two dimensions of our
model. “Causal ambiguity” and “lack of absorptive capacity” constitute as “origins
of stickiness” the key barriers explaining the stickiness of practices consulting com-
panies claim to be able to transfer (Barney, 1991; Grant, 1996; Decarolis and Deeds,
1999). The stickiness in turn manifests in the hampered ability of consultancies to
communicate and document such knowledge.
Efforts to overcome these barriers cannot be considered promising since compa-
nies are not keen to let consultancies analyze practices that supposedly constitute a
competitive advantage. In addition, existing studies show that implementation of
Table 7 Canonical weights for the two analyzed dimensions
Root
First Second Third
Origins of stickiness
Unproven knowledge 0.13 0.85 0.84
Causal ambiguity 0.44 1.08 0.45
Lack of absorptive capacity 0.61 0.38 1.00
Stickiness
Sticky documented BP 0.58 0.60 0.85
Sticky communication on BP 0.59 0.80 0.56
Barren context of BP 0.02 0.63 0.86
708 B. Wellstein and A. Kieser
at University Mannheim on March 27, 2015http://icc.oxfordjournals.org/Downloaded from

practices even in the same context struggles with severe transfer impediments. As, for
example, Newell (2005: 61) shows for the healthcare sector: “The new work prac-
tice ...could not be directly transferred to other hospital contexts where the pro-
posed recipients had not been through this change process.” Indeed, best practices
prove sticky, even for consultants.
8. Discussion
Consultancies’ claim that they are capable of identifying “best practices” is a myth.
Rather, best practice is a buzzword launched by consultants out of marketing pur-
poses. However, we cannot rule out that consultancies are able to identify practices
or concepts that are at least more effective than practices that are applied in many
client companies. Lean production may serve as an example. Interestingly, consult-
ants do not agree on the methods that help to identify effective practices. They are
not in a better position than management scholars.
What can be done to advance the discussion on transfer of practices? We think
that a definition of practices as bundles of rules and routines that can be interpreted
theoretically in a meaningful way might provide a better approach to identifying,
describing, learning about and, finally, transferring effective practices. In organiza-
tions, most activities are based on formal rules (standard operating procedures) and
routines that evolve around them when organizational members make sense of rules
and interpret what they have to do in order to bring the rules to life (March, 1994;
Kieser et al., 2001; Schulz and Jobe, 2001; Feldman and Rafaeli, 2002; Becker, 2004).
It is possible to record rules and the routines in existing organizations though the
recording is never complete and the transfer is therefore sticky. It is essential to
identify the basic mechanisms underlying those practices. Through careful imple-
mentation they may eventually become part of an organization’s life.
To identify the relatively more effective practices on the basis of statistical analyses
is, as we have pointed out in our conceptualization, extremely problematic. It is
therefore important to develop a theoretical model that contains assumptions on
the ways in which practices influence performance, including the risks that are
associated with it. This model should at least be plausible (Kaplan, 2003).
p < 0.0001
Stickiness
Sticky Documented BP
Sticky Communication on BP
Barren Context of BP
.58
.59
-.02
Origins of
Stickiness
Unproven Knowledge
Causal Ambiguity
Lack of Absorptive Capacity
.13
.44
.61
Figure 4 Results of canonical correlation.
Trading “best practices”—a good practice? 709
at University Mannheim on March 27, 2015http://icc.oxfordjournals.org/Downloaded from

Considering the fate of success factor research the hope that practices will definitely
be identified as best ones is futile. However, plausibility may be sufficient for the
practice of consultancies.
Special attention has to be paid to the phenomenon of stickiness. Consultants and
their clients should be aware—or made aware—that each transfer of a practice is an
experiment. The risk of the experiment may increase as differences increase between
the contexts of the organization in which a practice was identified and documented
and the contexts of the receiving organization. It is also necessary to carefully observe
the process of embedding—or rejecting—a new practice. This may take more time
after the actual implementation effort than a client is willing to pay a consultant for.
The client organization has to develop absorptive capacity, which includes a capacity
for rejecting nonfitting and absorbing practices that have been subject to a transfer
attempt by consultants.
9. Conclusion
Before we come to conclusions, several limitations of our study warrant mentioning.
One first restriction stems from the national character of our samples. However, as
big consulting companies act internationally, our study could be a springboard to
extend this stream of research further into an international realm. Another limitation
is that practices are implemented on an organizational level—different members
contribute to their embedding and collaborate in executing them. However, we
used—due to the lack of objective indicators—primarily the perceptions of consult-
ants on an individual level to measure constructs. As our analyses dealt with the
consultants’ point of view an avenue for further research could be to analyze data
that reflect the clients’ perceptions of the acquisition and implementation of best
practice knowledge through consulting projects.
We conclude that the most important finding from this study is the fragility of the
best practice concept. Both, from a theoretical as well as an empirical point of view,
“best practices” do not live up to the hopes that this label triggers. From a theoretical
point of view, neither benchmarking nor success factor research is able to identify
best practices. Our empirical analyses have shown that some consultants are well
aware of the limitations of attempts to identify and transfer them. We discovered
fundamental disagreement between consultants whether practices can be transferred
across industries or even between organizations. Due to factors like lack of absorptive
capacity and causal ambiguity stickiness is a problem that plagues attempts to trans-
fer them. These findings have important implications for research and practice alike.
Management research should be aware of the problems linked to the concept of
“best practices.” One can find many scientific articles in which this expression is used
in a non-reflective way. These studies do, for example, not take the concept
of stickiness into consideration nor do they adequately define their concept of
710 B. Wellstein and A. Kieser
at University Mannheim on March 27, 2015http://icc.oxfordjournals.org/Downloaded from

“best practice.” With our research, we extend extant insights into the identification
and transfer of best practices into the realm of companies that specialize on the
identification and transfer of successful practices. As our results suggest, scholars
should rethink what is needed to identify and transfer an effective practice.
Consultants position themselves as masterminds in pursuing such operation.
When scientists engage in studies of successful practices they should not compete
with consultants by imitating them but should pursue more modest aims and should
consider rules and routines as components of practices.
With our contribution we hope to make practitioners (more) aware of limitations
of acquiring such expertise. Management consultants do not possess the alchemy for
identifying and transferring best practices. They struggle with causality and absorp-
tion of “best practices.” When practices are seen as rules and routines it is clear for
the implanting company that rules and routines have to be developed. This process
takes time and is probably completed long after the consultants have left.
References
Accenture (2008), ‘Chem firms need to share SCM best practices,’ Purchasing, 137(9), 44.
American Diabetes Association (2005), ‘Standards of medical care in diabetes,’ Diabetes Care,
28(Suppl 1)S4–S36.
Anand, N., H. K. Gardner and T. Morris (2007), ‘Knowledge-based innovation: emergence
and embedding of new practice areas in management consulting firms,’ Academy of
Management Journal, 50, 406–428.
Armstrong, J. S. and T. S. Overton (1977), ‘Estimating nonresponse bias in mail surveys,’
Journal of Marketing Research, 14, 396–402.
Arthur, D. L. (2009), Press Release: Arthur D. Little Appoints MD for Middle East Region,
http://www.ameinfo.com/109252.html (22 March 2011, date last accessed).
Bagozzi, R. and Y. Yi (1988), ‘On the evaluation of structural equation models,’ Journal of the
Academy of Marketing Science, 16(1), 74–94.
Barney, J. (1991), ‘Firm resources and sustained competitive advantage,’ Journal of
Management, 17, 99–120.
Becker, M. C. (2004), ‘Organizational routines: a review of the literature,’ Industrial and
Corporate Change, 13, 643–677.
Bendell, T., I. Boulter and J. Kelly (1998), Benchmarking for Competitive Advantage. Pitman
Publishing: London.
Berta, W. B. and R. Baker (2004), ‘Factors that impact the transfer and retention
of best practices for reducing error in hospitals,’ Health Care Management Revue, 29(2),
90–97.
Bijmolt, T. H. A. and R. G. M. Pieters (2001), ‘Meta-Analysis in marketing when studies
contain multiple measurements,’ Marketing Letters, 12, 157–169.
Trading “best practices”—a good practice? 711
at University Mannheim on March 27, 2015http://icc.oxfordjournals.org/Downloaded from

Billesbach, T. J. (1994), ‘Applying lean production principles to a process facility,’ Production
& Inventory Management Journal, 35(3), 40–44.
Blascovich, J. D. (2005), ‘New A.T. Kearney study identifies emerging trends & best practices
in procurement,’ Supplier Selection & Management Report, 5(5), 2–5.
Boston Consulting Group (2009), ‘Boston Consulting Group, People’s Republic of
China Website Home Page,’ http://www.bcg.com.cn/en/default.html (1 March 2010, date
last accessed).
Bretschneider, S., F. Marc-Aurele Jr and J. Wu (2005), ‘“Best practices” research: a methodo-
logical guide for the perplexed,’ Journal of Public Administrative Research and Theory, 15,
307–323.
Burke, M. J. and W. P. Dunlap (2002), ‘Estimating interrater agreement with the average
deviation index: A user’s guide,’ Organizational Research Methods, 5(2), 159–172.
Camerer, C. and L. Fahey (1988), ‘The regression paradigm: A critical appraisal and suggested
directions,’ in J. H. Grant (ed.), Strategic Management Frontiers. JAI Press: Greenwich, UK,
pp. 443–459.
Camp, R. C. (1995), Benchmarking - The Search for Industry Best Practices that Lead to Superior
Performance. ASQC Quality Press: New York.
Chatterjee, S. and B. Price (1977), Regression Analysis by Example. Wiley: New York.
Churchill, G. A. (1979), ‘A paradigm for developing better measures of marketing constructs,’
Journal of Marketing Research, 16(1), 64–73.
Cohen, W. M. and D. A. Levinthal (1990), ‘Absorptive capacity: a new perspective on learning
and innovation,’ Administrative Science Quarterly, 35, 128–152.
Cyert, R. M. and J. G. March (1963), A Behavioral Theory of the Firm. Prentice Hall:
Englewood Cliffs, NJ.
Daft, R. L. and V. L. Buenger (1990), ‘Hitching a ride on a fast train to nowhere,’
in J. W. Frederickson (ed.), Perspectives on Strategic Management. Harper Business: Grand
Rapids, pp. 81–103.
Dani, S., J. A. Harding, K. Case, R. I. M. Young, S. Cochrane, J. Gao and D. Baxter (2006),
‘A methodology for best practice knowledge management,’ Proceedings of the Institution of
Mechanical Engineers – Part B – Engineering Manufacture, 220(10), 1717–1728.
Davies, A. J. and A. K. Kochhar (2002), ‘Manufacturing best practice and performance
studies: a critique,’ International Journal of Operations & Production Management , 22(3),
289–305.
Decarolis, D. M. and D. L. Deeds (1999), ‘The impact of stocks and flows of organizational
knowledge on firm performance: an empirical investigation in the biotechnology industry,’
Strategic Management Journal, 20(10), 953–969.
Denrell, J. (2005), ‘Selection bias and the perils of benchmarking,’ Harvard Business Review,
83(4), 114–119.
Denzin, N. K. (1978), The Research Act: A Theoretical Introduction to Sociological Methods.
McGraw Hill: New York.
712 B. Wellstein and A. Kieser
at University Mannheim on March 27, 2015http://icc.oxfordjournals.org/Downloaded from

Dess, G. G. and P. S. Davis (1984), ‘Porter’s (1980) generic strategies as determinants of
strategic group membership and organizations performance,’ Academy of Management
Journal, 27, 467–488.
Dimancescu, D., P. Hines and N. Rich (1997), The Lean Enterprise: Designing and Managing
Strategic Processes for Customer Winning Performance. Amacom: New York.
Dooley, L. M. and J. R. Lindner (2003), ‘The handling of nonresponse error,’ Human Resource
Development Quarterly, 14(1), 99–110.
Dosi, G., D. J. Teece and S. G. Winter (1992), ‘Toward a theory of corporate coherence: some
preliminary remarks,’ in G. Dosi, R. Giannetti and P. A. Toninelli (eds), Technology and
Enterprise in a Historical Perspective. Clarendon Press: Oxford, pp. 185–211.
Edmondson, A. C., R. M. Bohmer and G. P. Pisano (2001), ‘Disrupted routines: team learning
and new technology implementation in hospitals,’ Administration Science Quarterly, 46,
685–716.
Ericsson, K. and H. Simon (1980), ‘Verbal reports as data,’ Psychological Review, 87 (3),
215–251.
Ericsson, K. and H. Simon (1993), Protocol Analysis: Verbal Reports as Data. MIT Press:
Cambridge, MA.
Feldman, M. S. and A. Rafaeli (2002), ‘Organizational routines as sources of connections and
understandings,’ Journal of Management Studies, 39, 309–332.
Flynn, B. B., S. Sakakibara and R. G. Schroeder (1995), ‘Relationship between JIT and TQM:
practices and performance,’ Academy of Management Journal, 38(5), 1325–1360.
Fornell, C. and D. F. Larcker (1981), ‘Evaluation structural equation models with
unobservable variables and measurement error,’ Journal of Marketing Research, 18(1),
39–50.
Francis, G. and J. Holloway (2007), ‘What have we learned? Themes from the literature on
best-practice benchmarking,’ International Journal of Management Reviews, 9(3), 171–189.
Ghemawat, P. (1991), Commitment: The Dynamic of Strategy. Free Press: New York.
Grant, R. M. (1996), ‘Toward a knowledge-based theory of the firm,’ Strategic Management
Journal, 17, 109–122.
Hambrick, D. C. (2007), ‘The adolescence of strategic management, 1980-1985: critical per-
ceptions and reality,’ in I. A. Ansoff (ed.), Strategic Management. Grand Rapids: Harper
Business, pp. 464–477.
Hodgson, G. M. (1998), ‘The approach of institutional economics,’ Journal of Economic
Literature, 36, 663–684.
Holloway, J., G. Francis and M. Hitnon (1999), ‘A vehicle for change? A case study of per-
formance improvement in the “new” public sector,’ The International Journey of Public
Sector Management, 12, 351–364.
Huselid, M. A., S. E. Jackson and R. S. Schuler (1997), ‘Technical and strategic human re-
sources management effectiveness as determinants of firm performance,’ Academy of
Management Journal, 40, 171–188.
Trading “best practices”—a good practice? 713
at University Mannheim on March 27, 2015http://icc.oxfordjournals.org/Downloaded from

Jensen, R. J. and G. Szulanski (2007), ‘Template use and the effectiveness of knowledge trans-
fer,’ Management Science, 53, 1716–1730.
Jensen, R. and G. Szulanski (2004), ‘Stickiness and the adaptation of organizational prac-
tices in cross-border knowledge transfers,’ Journal of International Business Studies, 35,
508–523.
Kaplan, S. (2003), ‘The seduction of best practice: Commentary on “Taking strategy serious-
ly”,’ Journal of Management Inquiry, 12, 410–413.
Kennedy, P. (1998), A Guide to Econometrics. MIT Press: Cambridge.
Kieser, A., N. Beck and R. Tainio (2001), ‘Rules and organizational learning. The behavioral
approach,’ in J. Child, M. Dierkes and A. Berthoin Antal (eds), Handbook of Organizational
Learning. Oxford University Press: London, pp. 598–623.
Kieser, A. and U. Koch (2008), ‘Can organizations learn without their members learning from
each other?’ Management Learning, 39, 329–347.
Kieser, A. and A. Nicolai (2005), ‘Success factor research: overcoming the trade-off betwen
rigor and relevance?’ Journal of Management Inquiry , 14, 275–279.
King, A. W. and C. P. Zeithaml (2003), ‘Measuring organizational knowledge: a conceptual
and methodological framework,’ Strategic Management Journal, 24, 763–772.
Krafcik, J. F. (1988), ‘Triumph of the lean production System,’ Sloan Management Review,
30(1), 41–52.
Lampel, J. and Z. Shapira (1995), ‘Progress and its discontents: data scarcity and the limits
of falsification in strategic management,’ Advances in Strategic Management, 12A, 113–150.
Liker, J. K., W. M. Fruin and P. S. Adler (1999), Remade in America: Transplanting and
Transforming Japanese Management Systems. Oxford University Press: New York.
Lippmann, S. and R. P. Rumelt (1982), ‘Uncertain imitability: an analysis of interfirm differ-
ences in efficiency under competition,’ Bell Journal of Economics, 12, 418–438.
Mahoney, J. T. and J. R. Pandian (1992), ‘The resource-based view within the conversation of
strategic management,’ Strategic Management Journal, 13, 363–380.
March, J. G. (1994), A Primer on Decision Making: How Decisions Happen. The Free Press:
New York.
March, J. G., M. Schulz and X. Zhou (2000), The Dynamics of Rules: Change in Written
Organizational Codes. Stanford University Press: Stanford.
March, J. G. and R. I. Sutton (1997), ‘Organizational performance as a dependent variable,’
Organization Science, 6, 698–706.
Matzler, K., H. Hinterhuber, C. Daxer and M. Huber (2005), ‘The relationship between cus-
tomer satisfaction and shareholder value,’ Total Quality Management & Business Excellence,
16(5), 671–680.
McAlister, F. A., S. van Diepgen, R. S. Padwal, J. A. Johnson and S. R. Majumdar (2007), ‘How
evidence-based are the recommendations in evidence-based guidelines?’ Public Library of
Science Medicine, 4(8), 1–14.
714 B. Wellstein and A. Kieser
at University Mannheim on March 27, 2015http://icc.oxfordjournals.org/Downloaded from

Meyer, A. D. (1991), ‘What is strategy’s distinctive competence?’ Journal of Management, 17,
821–833.
Meyer, T. W. (1994), ‘Measuring performance in economic organizations,’ in N. J. Smelser
and R. Swedberg (eds), Handbook of Economic Sociology. Princeton University Press:
Princeton, NJ, pp. 556–588.
Miles, M. B. and A. M. Huberman (1994), Qualitative Data Analysis: An Extended Sourcebook
of New Methods. Sage: Thousand Oaks, CA.
Miner, A. S. (1994), ‘Seeking adaptive advantage: evolutionary theory and managerial action,’
in J. A. C. Baum and J. V. Singh (eds), Evolutionary Dynamics of Organizations. Oxford
University Press: New York, pp. 76–89.
Mintzberg, H. (1994), The Rise and Fall of Strategic Planning. The Free Press: New York.
Morris, S. S., P. M. Wright, J. Trevor, P. Stiles, G. K. Stahl, S. Snell, J. Paauwe and E. Farndale
(2009), ‘Global challenges to replicating HR: the role of people, processes, and systems,’
Human Resource Management, 48, 973–995.
Nattermann, P. M. (2000), ‘Best practice 6¼best strategy,’ McKinsey Quarterly, 2, 22–31.
Nelson, R. R. and S. G. Winter (1982), An Evolutionary Theory of Economic Change. Harvard
University Press: Cambridge.
Newell, S. (2005), ‘The falacy of simplistic notions of the transfer of “best practice”,’
in A. F. Buono and F. Poulfelt (eds), Challenges and Issues in Knowledge Management.
Information Age Publication: Greewich, CT, pp. 51–68.
Orlikowski, W. J. (2002), ‘Knowing in practice: enacting a collective capability in distributed
organizing,’ Organization Science, 13, 249–273.
Palich, L. E., L. B. Cardinal and C. C. Miller (2000), ‘Curvilinearity in the diversification-
performance linkage: an example of over three decades of research,’ Strategic Management
Journal, 23, 155–174.
Ray, G., J. B. Barney and W. A. Muhanna (2004), ‘Capabilities, business processes, and com-
petitive advantage: choosing the dependent variable in empirical tests of the resource-based
view’,’ Strategic Management Journal, 25(1), 23–37.
Reed, R. and R. J. DeFillippi (1990), ‘Causal ambiguity, barriers to imitation, and sustainable
competitive advantage,’ Academy of Management Review, 15(1), 88–102.
Rinehart, J., C. Huxley and D. Robertson (1997), Just Another Car Factory? Lean Production
and Its Discontents. Cornell University Press: Ithaca.
Rogelberg, S. G., J. M. Conway and M. E. Sederburg (2003), ‘Profiling active and passive
nonrespondents to an organizational survey,’ Journal of Applied Psychology, 88(6),
1104–1114.
Rousseau, D. M., J. Manning and D. Denyer (2008), ‘Evidence in management and organiza-
tional science: assembling the field’s full weight of scientific knowledge through syntheses,’
The Academy of Management Annals, 2(1), 475–515.
Sambharya, R. B. (2000), ‘Assessing the construct validity of strategic and SIC-based measures
of corporate diversification,’ British Journal of Management, 11(2), 163–173.
Trading “best practices”—a good practice? 715
at University Mannheim on March 27, 2015http://icc.oxfordjournals.org/Downloaded from

Sanwal, A. (2008), ‘The myth of best practices,’ Journal of Corporate Accounting & Finance,
19(5), 51–60.
Scarbrough, H. and M. Terry (1998), ‘Forget Japan: the very British response to lean produc-
tion,’ Employee Relations, 20, 224–236.
Schulz, M. and L. A. Jobe (2001), ‘Codification and tacitness as knowledge management
strategies: an empirical exploration,’ Journal of High Technology Management Research,
12, 139–165.
Sewell, G. and B. Wikinson (1992), ‘Someone to watch over me: surveillence, discipline and
the JIT labour process,’ Sociology, 26, 271–289.
Shaneyfelt, T. M., M. F. Mayo-Smith and J. Bothwangl (1999), ‘Are guidelines following
guidelines? The methodological quality of clinical practice guidelines in the peer-reviewed
medical literature,’ Journal of the American Medical Association, 281, 1900–1905.
Sherry, A. and R. K. Henson (2005), ‘Conducting and interpreting canonical correlation
analysis in personality reserach: a user-friendly primer,’ Journal of Personality Assessment,
84(1), 37–48.
Simonin, B. L. (1999), ‘Ambiguity and the process of knowledge transfer in strategic alliances,’
Strategic Management Journal, 20, 595–624.
Starbuck, W. H. (1993), ‘Keeping a butterfly and an elephant in a house of cards: the elements
of exceptional success,’ Journal of Management Studies, 30, 885–921.
Strack, F. and L. L. Martin (1987), ‘Thinking, judging and communicating: a process account
of context effects in attitude surveys,’ in H. J. Hippler, N. Schwarz and S. Sudman (eds),
Social Information Processing and Survey Methodology. McGraw Hill: New York, pp. 47–68.
Szulanski, G. (1996), ‘Exploring internal stickiness: impediments to the transfer of best pratice
within the firm,’ Strategic Management Journal, 17(S2), 27–43.
Szulanski, G. and R. Cappetta (2003), ‘Stickiness: conceptualizing, measuring, and predicting
difficulties in the transfer of knowledge within organizations,’ in M. Easterby-Smith and
M. L. Lyles (eds), The Blackwell Handbook of Organizational Learning and Knowledge
Management. Blackwell: Oxford, pp. 513–534.
Szulanski, G., R. Cappetta and R. J. Jensen (2004), ‘When and how thrustworthiness matters:
knowledge transfer and moderating effect of causal ambiguity,’ Organization Science, 15(5),
600–613.
Szulanski, G. and R. J. Jensen (2008), ‘Growing through copying: the negative consequences of
innovation on franchise network growth,’ Research Policy, 37(10), 1732–1741.
Thorndike, R. M. (2000), ‘Canonical correlation analysis,’ in H. E. A. Tinsley and S. D. Brown
(eds), Handbook of Applied Multivariate Statistics and Mathematical Modeling. Academic
Press: San Diego, CA, pp. 237–263.
Tucker, A. L., I. M. Nembhard and A. C. Edmondson (2007), ‘Implementing new practices: an
empirical study of organizational learning in hospital intensive care units,’ Management
Science, 53(6), 894–907.
716 B. Wellstein and A. Kieser
at University Mannheim on March 27, 2015http://icc.oxfordjournals.org/Downloaded from

Wailes, N. and G. Michelson (2008), ‘The transfer of management ideas to a Western “per-
iphery”: The case of corporate social responsibility in Australia,’ International Studies of
Management & Organization, 38(4), 100–118.
Walgenbach, P. and C. Hegele (2001), ‘What can an apple learn from an orange? Or: What do
companies use benchmarking for?’ Organizations, 8, 121–144.
Wareham, J. and H. Gerrits (1999), ‘De-contextualising competence: can business best practice
be bundled and sold?’ European Management Journal, 17, 39–49.
Williams, K., C. Haslam, J. Williams, T. Cutler, A. Adcroft and S. Johal (1992), ‘Against lean
production,’ Economy and Society, 21, 321–354.
Winter, S. G. (1964), ‘Economic “natural selection” and the theory of the firm,’ Yale Economic
Essays, 4, 225–272.
Womack, J. P., D. T. Jones and D. Ross (1990), The Machine that Changed the World. Rawson
Associates: New York.
Woywode, M. (2002), ‘Global manufacturing concepts and local adaptation: working groups
in the French and German car manufacturing industry,’ Organization Studies , 23, 497–524.
Zollo, M. and S. G. Winter (2002), ‘Deliberate learning and the evolution of dynamic cap-
abilities,’ Organization Science, 13, 339–351.
Trading “best practices”—a good practice? 717
at University Mannheim on March 27, 2015http://icc.oxfordjournals.org/Downloaded from

Appendix A: Measures and items (Scales 1–7 converted to 0–100)
Table A1 Origins of stickiness
718 B. Wellstein and A. Kieser
at University Mannheim on March 27, 2015http://icc.oxfordjournals.org/Downloaded from

Table A2 Stickiness
Trading “best practices”—a good practice? 719
at University Mannheim on March 27, 2015http://icc.oxfordjournals.org/Downloaded from