1
.
.
.
.
.
.
.
.
.
COMMUNICATING AND CO NTROLLING STRATEGY:
AN EMPIRICAL STUDY OF THE EFFECTIVENESS OF
THE BALANCED SCORECARD
Mary A. Malina
University of Melbourne
and
Frank H. Selto
University of Colorado at Boulder
University of Melbourne
APPEARS IN JOURNAL OF MANAGEMENT ACCOUNTING RESEARCH, V. 13, 2001
We gratefully acknowledge helpful comments from Luke George, David Guenther, Marlys Lipe,
Peter Luckett, Bill Maguire, Axel Schulz, Phil Shane, Naomi Soderstrom, and Wim Van Der Stede,
workshop participants at the University of Colorado, University of Melbourne, AAANZ 2000 and
AAA 2000, and, in particular, the two anonymous reviewers who gave consistently insightful and
constructive comments. This research was supported by a Hart Doctoral Fellowship from the
University of Colorado at Boulder and by data generously provided by the anonymous host
company.
1
COMMUNICATING AND CONTROLLING STRATEGY:
AN EMPIRICAL STUDY OF THE EFFECTIVENESS OF
THE BALANCED SCORECARD
ABSTRACT
This paper reports evidence on the effectiveness of the Balanced Scorecard (BSC) as a strategy
communication and management-control device. This study first reviews communication and management
control literatures that identify attributes of effective communication and control of strategy. Second, the
study offers a model of communication and control applicable to the BSC. The study then analyzes empirical
interview and archival data to model the use and assess the communication and control effectiveness of the
BSC. The study includes data from multiple divisions of a large, international manufacturing company. Data
are from BSC designers, administrators, and North American managers whose divisions are objects of the
BSC. The study accumulates evidence regarding the challenges of designing and implementing the BSC faced
by even a large, well-funded company. These findings may be generalizable to other companies adopting or
considering adopting the BSC as a strategic and management control device.
Data indicate that this specific BSC, as designed and implemented, is an effective device for controlling
corporate strategy. Results also indicate disagreement and tension between top and middle management
regarding the appropriateness of specific aspects of the BSC as a communication, control and evaluation
mechanism. Specific results include evidence of causal relations between effective management control,
motivation, strategic alignment and beneficial effects of the BSC. These beneficial effects include changes in
processes and improvements in both the BSC and customer-oriented services. In contrast, ineffective
communication and management control cause poor motivation and conflict over the use of the BSC as an
evaluation device.
Data availability: Use of all data collected for this study is regulated by a strict non-disclosure agreement,
which requires the researchers to protect the company’s identity and its proprietary information.
2
COMMUNICATING AND CONTROLLING STRATEGY:
AN EMPIRICAL STUDY OF THE EFFECTIVENESS OF
THE BALANCED SCORECARD
INTRODUCTION
The professional and academic strategy literatures claim that many organizations have found traditional
performance measures (e.g., ex post costs, profits, and return on investment) to be insufficient guides for
decision making in today’s rapidly changing, hyper-competitive environment. Sole reliance on current,
financial measures of performance arguably does not reflect the importance of current resource decisions for
future financial performance [e.g., Dearden, 1969]. Though some firms recognized the importance of non-
financial measures of performance many years ago (e.g., General Electric in the 1950s), growing international
competition and the rise of the TQM movement have widened the appeal of non-financial performance
measures. Since the 1980s, authors have filled the professional and academic literature with recommendations
to rely more on non-financial measures for both managing and evaluating organizations [e.g., Johnson and
Kaplan, 1987; Berliner and Brimson, 1988; Nanni et al., 1988; Dixon et al., 1990; Rappaport, 1999].
In addition to normative arguments, empirical research can help to establish the roles and effectiveness
of non-financial performance measurement. A number of studies have sought to link specific non-financial
measures to financial performance (e.g., Banker et al., 2000; Behn and Riley, 1999; Foster and Gupta, 1999;
Ittner and Larcker, 1998a).
1
Evidence in the human resources literature shows that systems of non-financial
measures, not individual measures themselves, appear to be more reliable determinants of firm performance.
(e.g., Becker and Huselid, 1998; Huselid, 1995, 1997). The objective of this study is to examine the process
and impact of managing an organization with non-financial performance measures, specifically in the context
of the balanced scorecard (BSC), which is a comprehensive system of performance measurement.
The BSC, popularized by Kaplan and Norton [1992, 1993, 1996a, b, c] and adopted widely around the
world, has been offered as a superior combination of non-financial and financial measures of performance.
2
Because the BSC explicitly focuses on links among business decisions and outcomes, it is intended to guide
strategy development, implementation, and communication. Furthermore, a properly constructed BSC could
provide reliable feedback for management control and performance evaluation.
Atkinson et al. [1997] regard the BSC as one of the most significant developments in management
accounting, deserving intense research attention. Silk [1998] estimated that 60 percent of the U.S. FORTUNE
500 companies have implemented or are experimenting with a BSC. Given its high profile, surprisingly little
academic research has focused on either the claims or outcomes of the BSC [Ittner and Larcker, 1998b]. A
natural question is: does the BSC’s content, format, implementation, or use have discernable effects on
3
business decisions and outcomes that could not be attained with existing measures, alone or in combination?
In the first study of its kind, Lipe and Salterio [2000] identify decision effects associated with the format of
the BSC. The arrangement of performance measures into four related categories appears to convey decision-
relevant information to subjects performing a laboratory evaluation task. Most other current BSC studies,
however, are relatively uncritical descriptions of BSC adoptions.
Kaplan and Norton [1996] argue that the BSC is not primarily an evaluation method, but is a strategic
planning and communication device to (1) provide strategic guidance to divisional managers and (2) describe
links among lagging and leading measures of financial and non-financial performance. The BSC purports to
describe the steps necessary to reach financial success; for example, invest in specific types of knowledge to
improve processes. If the links are valid reflections of a company’s administrative and productive processes
and economic opportunities, the BSC embodies and can communicate the company’s operational strategy.
Furthermore, effectively communicating these links throughout the organization can be crucial to
implementing that strategy successfully [Tucker et al, 1996; West and Meyer, 1997]. Organizations also might
use non-financial measures as the basis of performance evaluation. Alternatively, they might improve
performance by using the BSC as a guide to financial success and by also judicially using financial
performance measures for evaluation purposes [e.g., Rappaport; 1999].
The present study investigates the communication and management-control attributes and the
effectiveness of a large, successful, international company’s BSC model. The study includes archival and
qualitative data from interviews with the BSC’s designers, managers, and users to (1) assess the perceived
attributes of the BSC as both a strategic communication and control device and (2) find evidence of the
BSC’s decision impacts. The current study does not test whether the company’s BSC is a statistically valid
model of the company’s activities and performance. This feature of the BSC will be tested in subsequent
research [Malina, 2001].
The company introduced the BSC to advance its strategy. The scorecard has greatly affected the outlook
and actions of users, both beneficially and adversely. When elements of the BSC are well designed and
effectively communicated (according to criteria described in the study), the BSC appears to motivate and
influence lower-level managers to conform their actions to company strategy. Furthermore, managers believe
that these changes result in improved sub-unit performance. However, there also is consistent evidence that
flaws in the BSC design and shortcomings in strategic communication have adversely affected relations
between some top and middle managers. The tension exists because the BSC design exacerbates strong
differences between their views of future opportunities. Shortcomings in communication generate mistrust
and unwillingness to change. While the specific flaws and shortcomings may be unique to the studied
company, these findings appear to reflect generally on issues of BSC design and uses.
4
The second section of this paper develops a research question from a review of the communication
literature regarding characteristics of effective communication of strategy. The third section develops a
second research question through an overview of attributes of management control devices that effectively
control strategy. The fourth section describes the research site and the company’s BSC. The fifth section
describes procedures used to obtain and analyze the archival and qualitative interview data. This section also
presents a theoretical model to describe BSC effectiveness. The sixth section addresses the research questions
and derives an empirical model of BSC effectiveness. The final section summarizes conclusions and offers
recommendations for future research.
THE BSC AND COMMUNICATION OF STRATEGY
Kaplan and Norton [1996c] state that, “by articulating the outcomes the organization desires as well as
the drivers of those outcomes (by using the BSC), senior executives can channel the energies, the abilities, and
the specific knowledge held by people throughout the organization towards achieving the business’s long-
term goals.” Thus, Kaplan and Norton assert that not only does the BSC embody or help create
organizational strategy and knowledge, but also the BSC itself effectively communicates strategy and
knowledge. Merchant [1989] argues that communication failure is an important cause of poor organizational
performance. Because no organization’s knowledge or strategy exists apart from or succeeds without its key
human actors, the ability to effectively communicate may be itself a source of competitive advantage [Tucker
et al., 1996; Daft and Lewin, 1993; Grant, 1991; Schulze, 1992; Amit and Shoemaker, 1990]. If the BSC does
articulate organizational knowledge and strategy in a superior manner, then it may be a source of competitive
advantage, at least until all competitors use it equally well. The organizational communication literature,
however, identifies a complex set of characteristics that affect the quality or effectiveness of communication
in organizations.
Based on a review of the literature, an organizational communication device or system may be
characterized by the attributes of its (1) processes and messages, (2) support of organizational culture, and (3)
creation and exchange of knowledge. Brief reviews of these communication characteristics follow.
Communication Processes and Messages
Individuals use and rely on communication if its processes and messages are perceived as understandable
and trustworthy. Other characteristics of effective organizational communication processes are routineness,
predictability, reliability, and completeness [Barker and Camarata, 1998; Goodman, 1998; Tucker, et al., 1996].
Communication also is more effective if it uses concise messages and clearly defined terms [Goodman, 1998].
Furthermore, an effective communication system precludes suppression of truth or misstatement of
performance. There should be no ambiguity regarding the differences between truthfulness and “looking
good” or integrity with winning. The effective communication system and its users will be intolerant of “spin,
5
deniability, and truth by assertion” [Goodman, 1998]. Therefore, organizational communication will be
effective if processes and messages are valid representations of performance. Effective communication and
effective performance measurement conceptually overlap, as was discussed previously.
Support of Culture, Values, and Beliefs
The traditional view of effective organizational communication is that it supports organizational culture
and individual interest by reinforcing desired patterns of behavior, shared values, and beliefs. Effective
communication demonstrates that the organization does what it says and that individual or group rewards are
predicated on their actions [Tucker, et al, 1996; Goodman, 1998]. Communication by leaders that consistently
articulates shared goals, values and beliefs [Tucker, et al, 1996; Goodman, 1998] also is effective in reinforcing
culture and directing behavior. Furthermore, effective communication must encourage behavior consistent
with organizational goals, values, and beliefs [Goodman, 1998].
Proponents of the BSC [e.g., Kaplan and Norton, 2000] argue that it also can be an instrument of cultural
and strategic change. Consistent with Kotter’s [1995] observations of change processes, the BSC may
facilitate change by effectively creating and communicating a credible vision of and method for achieving
change.
Creation and Exchange of Knowledge
Knowledge, which may be objective or tacit, is the basis of strategy formulation and implementation.
3
Therefore, an effective communication system supports an organization’s strategy by nurturing both objective
and tacit knowledge. The effective communication system exchanges objective (observable) knowledge
among key individuals so that all are aware of the organization’s current status. Organizations create objective
knowledge from the development and integration of new knowledge by individual specialists. Objective
knowledge usually derives from the refining and sharing of individuals’ tacit knowledge, which is understood
but not yet articulated or usable by the organization. Therefore, an effective communication system
encourages and enables the sharing of individuals’ experiences and collects those shared experiences. This
may be best accomplished by intense and frequent sharing, and by dialogue rather than one-directional
reporting. Perhaps importantly for the effectiveness of the BSC, de Haas and Kleingeld [1999] argue further
that participation in the design of performance measurement systems is an important determinant of effective
communication of strategy.
In summary, effective organizational communication devices should possess the observable attributes of
• Valid messages – reliable, understandable, trustworthy
• Support of organizational culture – existing or changing
• Knowledge-sharing – including dialogue and participation
6
The organizational communication literature predicts that a BSC, which has these attributes, will create
strategic alignment, positive motivation, and positive organizational outcomes. The first research question is:
1. Is the BSC an (in)effective communication device, creating strategic (non)alignment,
(in)effective motivation, and (negative)positive organizational outcomes?
THE BSC AND MANAGEMENT CONTROL OF STRATE GY
A common criticism of managing organizations based on financial measures of performance is that these
measures induce managers to make myopic, short-run decisions. Financial measures tend to focus on the
current impacts of decisions without a clear link between short-run actions and long-run strategy [recent
criticisms include McKenzie and Schilling, 1998; Luft & Shields, 1999]. Furthermore, traditional financial
measures of performance can work against knowledge-based strategies by treating the enhancement of
resources such as human capital, which may be critical to implementing strategy, as current expenses [e.g.,
Johnson, 1992]. Dixon et al. [1990] argue that traditional financial measures, by expensing costs of many
improvements, also work against strategies based on quality, flexibility, and minimization of manufacturing
time. For many lower-level employees, most financial performance measures are too aggregated and too far
removed from their actions to provide useful guidance or feedback on their decisions. They might need
measures that more directly and accurately relate to outcomes that they can influence [McKenzie and
Schilling, 1998]. A number of studies have found evidence that traditional, financial measures of performance
are most useful in conditions of relative certainty and low complexity – not the conditions faced by many
organizations today [e.g., Gordon and Naranyan, 1984; Govindarajan, 1984; Govindarajan and Gupta, 1985;
Abernethy and Brownell, 1997].
Lynch and Cross [1995] argue that performance measures should motivate behavior leading to
continuous improvement in key areas of competition, such as customer satisfaction, flexibility, and
productivity. That is, they should reflect cause and effect between operational behavior and strategic
outcomes [Keegan et al, 1989; Ittner and Larcker, 1998a].
4
Furthermore, as an organization identifies new
strategic objectives, it also may realize a need for new performance measures that encourage and monitor new
actions [Dixon et al., 1990]. Thus, organizations sensibly and perhaps optimally may use a diverse set of
performance measures to reflect the diversity of management decisions and efforts [e.g., Holmstrom, 1979;
Banker and Datar, 1989; Feltham and Xie, 1994; Ittner and Larcker, 1998b]. Empirical support for these
propositions is limited but growing.
5
The Management-Control Case for the Balanced Scorecard
Kaplan and Norton [1996b] have arranged multiple performance measures into the Balanced Scorecard,
which is a logical expression of most models of western business management.
6
Indeed, the BSC may have
7
spread widely throughout the world on the strength of its intuition and internal logic. Kaplan and Norton
claim that the BSC offers two significant improvements over traditional financial or even non-financial
measures of performance.
First, the BSC identifies four related areas of activity that may be critical to nearly all organizations and all
levels within organizations:
• Investing in learning and growth capabilities
• Improving efficiency of internal processes
• Providing customer value
• Increasing financial success
Following the logic of the BSC and ignoring cost-benefit considerations, most organizations could use
measures in all four areas to encourage and monitor actions appropriate to organizational strategy. In its most
basic use, a properly configured BSC could provide a comprehensive picture of the state of the organization,
much as an automobile’s dashboard shows fuel level, oil pressure, coolant temperature, engine RPM, and
velocity. Thus, the BSC might promote positive organizational outcomes such as improvements in all four
areas of organizational activity, which include administrative activities and the BSC itself. Assessing this first
level of effectiveness is the objective of this research.
Furthermore, the BSC seeks to link these measures into a model that accurately reflects cause and effect
relations among categories and individual measures. Using the automobile analogy, the BSC simulates a
change in a car’s performance (e.g., velocity) given a planned increase in fuel consumption and engine RPM
(and perhaps other factors). Such a model might support operational decisions, make predictions of
outcomes given decisions and environmental conditions, and provide reliable feedback for learning and
performance evaluation.
7
The Role of the BSC for Strategy Implementation and Performance Measurement
Proponents of the BSC stress its alignment of critical measures with strategy and links of the measures to
valued outcomes. In addition, the management control literature identifies other characteristics of control
systems that may be critical to the successful implementation of strategy and should apply to the BSC.
8
To be
effective, BSC measures should be accurate, objective, and verifiable. Otherwise, measures will not reflect
performance and may be manipulated, or managers could in good faith achieve good measured performance
but cause the organization harm. If managers can achieve good measured performance by cheating, the
system quickly will lose credibility and desired motivational effect. Furthermore, the set of BSC measures
should completely describe the organization’s critical performance variables, but should be limited in number
to keep the measurement system cognitively and administratively simple. An exhaustive set of performance
measures may accurately reflect the complexity of the organization’s tasks, but too many measures may be
8
distracting, confusing, and costly to administer. However, Lipe and Salterio [2000] did not find evidence of
information overload from multiple measures in their experimental study of the BSC.
Positive motivational impact induces managers to exert effort to achieve organizational goals. While
informative but not controllable performance measures may be important, positive motivation requires that
at least some of the BSC measures should reflect managers’ actions. For example, relative performance
evaluation (e.g., across similar business units), which can identify “influenceable” but not completely
controllable outcomes, may be an important component of the BSC [e.g., Antle and Demski, 1988], but it
may not be sufficient by itself. Extensive goal-setting literature confirms that performance should be keyed to
challenging but attainable targets [e.g., Locke and Latham, 1990]. Without such explicit BSC targets,
performance likely would be lower than could be reasonably achieved. Finally to build goal commitment, the
BSC should be linked with prompt and well-understood rewards and penalties. Rewards that are delayed,
uncertain, or ambiguous may be ineffective motivational devices.
Therefore, even though an organization’s BSC reflects its critical performance variables and links to
valued outcomes, it may fail as an effective management control device if it lacks other attributes. For
example, Ittner et al. [2000] found that subjectivity in a bank’s BSC led to both its having little beneficial
impact and the bank’s reversion to short-term financial measures of performance. To summarize, an effective
management control device, which is capable of promoting desired organizational outcomes, should have the
following, observable management control attributes to, first, attain strategic alignment:
• A comprehensive but parsimonious set of measures of critical performance variables, linked with
strategy
• Critical performance measures causally linked to valued organizational outcomes
• Effective – accurate, objective, and verifiable – performance measures
Second, to further promote positive motivation, an effective management control device should have
attributes of:
• Performance measures that reflect managers’ controllable actions and/or influenceable actions, e.g.,
measured by absolute and/or relative performance
• Performance targets or appropriate benchmarks that are challenging but attainable
• Performance measures that are related to meaningful rewards
Management control theory predicts that, if the BSC has these attributes, it is likely that the BSC will promote
strategic alignment and positive motivation and outcomes. Therefore, the second research question, which
parallels the first, is:
2. Is the BSC an (in)effective management control device, creating strategic (non)alignment,
(in)effective motivation, and (negative)positive organizational outcomes?
9
Subsequent discussions elaborate the details of a model that reflects the two research questions. This
model, based on the literature review, shows that the BSC’s management control and communication
characteristics generate outcomes by creating strategic alignment and motivation (or not). This study also
describes efforts to collect data on an implemented BSC’s management control and organizational
communication attributes, as well as evidence on the BSC’s effects on strategic alignment, motivation, and
organizational outcomes. It is bold to judge the effectiveness of the BSC against evidence from a single, non-
experimental BSC implementation. However, a thorough examination of a critical case can be instructive and
generalizable to theory [i.e., analytical generalization, Yin, 1994: 30-32], which in this case is that the BSC can
be an effective strategy communication and management control device.
RESEARCH SITE AND BSC CHARACTERISTICS
Overview of the Research Site
The research site is a U.S. FORTUNE 500 company with more than 25,000 employees and $6 billion
sales of durable products and post-sale services. The company is regarded as a long-term, well-managed
company. It is succeeding in highly competitive domestic and foreign markets, characterized by competition
among relatively few, very large, international companies. The company recently adopted a customer- and
quality-driven strategy to improve its competitiveness, and consequently perceived a need to expand its
management controls and performance management beyond traditional, financial measures. The company
began changing its performance measurement systems with a BSC that focuses on a very important part of
the company. One and a half years before the start of this study, the company began its implementation of a
Distributor-BSC (DBSC), for its 31 North American distributorships, which are responsible for a large share
of the company’s sales. The company has sufficient resources to assign BSC responsibilities to key staff that
are championing its continued development and implementation. These staff members have had formal BSC
training and are not using services of outside consultants. The DBSC was developed centrally and imposed
on the distribution channel, with little initial input from distributors themselves.
The company’s distributors in North America have primary responsibility for retail sales and service of
company products. Distributorships are organized by geographical area and may not sell other companies’
competing products. Although they are independently owned, individuals with employment experience in the
company currently lead 30 of the 31 distributorships. Distributors operate under renewable three-year
contracts with the company, which are based on realized and expected future performance.
9
The authors gained access to this company because of a family relationship between one of the authors
and executives of the company.
10
In this sense the field study is serendipitous, but the site is attractive on a
priori, objective grounds, and would have been a top candidate in a purposive sampling approach.
11
10
To summarize, the company has a long history of effective management control, extensive resources, and
a commitment to communicate its strategy to its distributors. Furthermore, early in the investigation
researchers perceived considerable tension and possible resistance to change among parties affected by the
DBSC, which, as Ahrens and Dent [1998] counsel, usually makes for an engaging study. Thus, the company
and its DBSC project are ideal for field study research on the balanced scorecard.
Overview of the DBSC
Purpose of the DBSC
In line with its new customer-driven strategy, the company recently changed its distribution strategy from
one of operational efficiency to managing long-term customer relations. Until the DBSC, the company had
evaluated formal distributor performance solely on financial performance and market share. Company
documents and literature show that staff personnel designed the DBSC, top-down without input from
distributors, to communicate the company’s new retail distribution strategy to its distributors. Company
documents state the purposes of the DBSC are to:
• Highlight areas within distributorships that need improvement to enhance customer relations
• Provide an objective set of criteria, consistent with the company’s new strategic initiatives, to
guide and measure total distributor performance.
These purposes fall well within the scope of the use of the BSC as envisioned by Kaplan and Norton.
However, administrators who developed and use the DBSC describe two additional objectives, which have
far reaching implications for managing the company’s distribution system:
• DBSC performance will be used as the starting point for the three-year contract review process.
• The DBSC is used for comparing and ranking distributorships and may be used for
performance-based compensation.
Because the DBSC includes many previously unevaluated areas of performance, it represents a dramatic
change in communication, interactions, and formal relations between the company and its distributors. In
particular, using the DBSC for distributor contract renewal and compensation added significant economic
incentives and created uncertainty regarding the impacts of DBSC performance.
Structure of the DBSC
The DBSC contains measures of performance in each of the four BSC perspectives plus another for
corporate citizenship, which the company felt was lacking in Kaplan and Norton’s specification of the BSC.
12
Additionally, the company has arranged its DBSC measures in categories that reflect its own priorities and
culture. Though distributors prepare some DBSC measures in “real time,” the company staff compiles,
analyzes, and disseminates the DBSC quarterly to top management and to distributors. An internal document
(usual BSC categories shown in brackets) describes the DBSC as:
11
“…comprised of measures that are categorized into groups which are aligned with [the company’s strategic]
objectives: Competitive Advantage [customer value and internal processes], Profitability and Growth [internal
processes and financial success], Corporate Citizenship, and Investments in Human Capital [learning and
growth]. A fifth category has been added to include other measures important to distributor performance
[internal processes]. Each of the categories includes specific measures with specific criteria for acceptability.
The results for the measures within each category will be weighted to determine an overall score for each
category and an overall score for the distributorship.”
A summary of the measures and the weights currently used in the DBSC are in table 1. For comparability
with the literature, we have arranged these measures into the usual BSC categories, but we also note where
the company has placed them in its own categories.
TABLE 1
DBSC Measures and Approximate Weights
Both distributors and DBSC administrators understood immediately that the DBSC’s relative weights
reflect the company’s view of the most important areas of performance.
13
Distributors’ knowledge of “why”
came later, if at all, as will be seen. Additionally and with experience, the company revised the weights to
reflect learning about measures’ impacts, reliability, or possible manipulation, particularly of some of the
softer measures, as Flamholtz [1979] predicts. One of the principal designers of the DBSC stated:
“Changes in weights are a function of two things: 1) how important we think the things are; 2) how credible the
numbers we get are…. How do we measure outstanding people at the distributor? It’s important, but how
substantive a measure can we come up with for it? Hardness of the number definitely affects the weights. If
we place a heavier weighting on something you don’t have confidence in, is that better than now?” [11:71-83].
One of the key managers of the distribution channel also explained:
“Now market share really is the driver and means more than the other things do. We did move more of the
weighting there. It’s more important to reflect the feeling of the management team. The distributors say, tell me
how you’re ranking me, and I’ll do it even if I don’t like it.” [12: 143-149]
The company’s first version of the DBSC placed a total of 20 percent weighting on Investments in human
capital (learning and growth area), but after a year that weight had been reduced to only 4 percent, primarily
because management felt the numbers were unreliable. Likewise, the first scorecard placed a 10 percent
weighting on Corporate citizenship (internal processes and customer value areas), later reduced to 4 percent. The
company redistributed the original weightings mostly to the traditional market share measure (an outcome of
building customer relationships), which grew in importance from 12 percent to 28 percent, to reflect the
paramount importance of building long-term customer relationships that result in market share. The
company also has added weight to quickly diagnosing and solving customer problems (internal processes
12
area), which grew from 2 percent to 10 percent in importance, to reflect the company’s belief about an
essential element for building customer relationships. As discussed later, the weightings and changes in
weighting affected distributors’ perceptions about both the “balance” in the DBSC and the truly important
measures of importance.
Management did not consider distributors to be partners in the process of developing the DBSC, which
reflected the company’s traditional, top-down approach to management. A more open, participative approach
to the development and use of the DBSC (one attribute of effective communication) could have had an
impact on distributors’ acceptance of the DBSC’s and their subsequent performance. Furthermore, the
company did not explicitly design the DBSC to be a “strategy map,” in Kaplan and Norton’s [2000]
terminology,
14
but let the measures and weights “speak for themselves” as key performance indicators. The
top-down and ambiguous nature of the communication may have impeded the immediacy and effectiveness
of the DBSC message. As will be demonstrated later, distributors had strong feelings on this, which can
explain adverse impacts of the DBSC.
Figure 1 shows a quarterly DBSC, as reported to management for several representative distributors. This
scorecard, which is based on numerical measures, is notable for several reasons. First, each distributor’s
quantified and internally benchmarked performance measure is labeled and colored “red” for “fails to meet
criteria for acceptability”, “yellow” for “meets criteria for acceptability”, or “green” for “exceeds criteria for
acceptability.”
15
The total score in the last column is computed by multiplying each measure’s numerical score
by the appropriate weights. Second, each distributor obtains its own report and its relative, numerical ranking
(e.g., 7
th
out of 31). Furthermore, names of distributors that achieve “green” ratings are posted on the
company’s intranet for all to see.
16
FIGURE 1
Representative Distributor BSC Ratings and
Scores
RESEARCH METHOD
This study investigates its research questions with qualitative, interview data obtained from individuals
directly involved with the company’s DBSC. Thus, the evidence is perceptual in nature and, while it ideally
reflects the “reality” of the impact of the DBSC, it also may reflect individuals’ and researchers’ biases in ways
that are not easily detectable. The study’s research method attempted to mitigate the effects of these
unknown biases. The research method is described below.
17
Sampling
Because the DBSC represents a dramatic change in distribution strategy from operational efficiency to
managing customer relationships by the company’s distributors, we sought and obtained direct commentary
13
from two DBSC designers, three managers who use it to evaluate distributors, and nine of the 31 distributors.
Because the research is interested in all facets of the DBSC, the scope of the distributor sample is limited to
those who consistently reported complete or nearly complete data. At the time of the study, these distributors
had the full six quarters’ experience with the DBSC. While added experience may continue to refine
perceptions, the sampled distributors represent the most experienced distributors available at the time of the
study. This selection may bias the analysis if more experienced distributors that also report more complete
data have systematically different perceptions than other distributors. Another source of bias may be
scorecard performance, which could influence perceptions of the DBSC; the sample included nine
distributors who reflected overall red, yellow, and green ratings. Of the distributors reporting complete data,
only one “green” distributor was available and three “red,” so the sample was filled out with five “yellow”
distributors. At the time of the interviews, overall there were 2 green, 19 yellow, and 10 red distributors. The
sample also reflected geographic dispersion – three western, two midwestern, two southern, one northeastern
US, and one Canadian distributor.
After analyzing the interviews, we feel confident that we have obtained a
full range of distributor responses. As the interviews proceeded, responses became repetitive. While
additional “green” distributor interviews would have been desirable, we feel they would be unlikely to
contribute additional insights.
18
Data Collection
The researchers obtained archival data (background and policy documents and quarterly DBSC scores)
from managers who administer the DBSC. All interview data were obtained via telephone in mid-1999 after
sponsoring managers informed designers, other managers, and all 31 distributors that the researchers were
conducting this study and may call them for input about the DBSC. Interviews lasted from 45 minutes to 75
minutes, depending entirely on how much an interviewee had to say. The study used a semi-structured
interview format and assured respondents of anonymity.
19
To avoid responses that could be artifacts of the interview process itself, the researchers deliberately did
not ask leading questions regarding management control or communication attributes of the DBSC or
questions directly related to the study’s research questions. While the study’s use of management control and
organizational communication theories represents a deductive approach to research and does guide later
analysis and model building, we were not confident that we had identified all relevant factors related to the
effectiveness of the DBSC. At this stage, we preferred to gather data more freely and let the respondents’
natural, undirected commentary support, deny, or extend the theories.
20
An important benefit of this
approach is that respondents may identify factors that affect the effectiveness of the DBSC other than those
anticipated by the study’s theory.
The researchers asked each distributor the following open questions:
1. In your own words, what is the distributor-balanced scorecard?
14
2. What do you think the objective of the balanced scorecard is?
3. What are the nine measures that distributors report really measuring?
4. What are the measures that are filled out by the company really measuring?
5. How do the measures that distributors report relate to the company’s measures? (Follow up: Do
changes in distributor performance cause changes in the company’s measures?)
6. Do the measures (distributors’ and the company’s) help you in any way? (Follow up: How?)
7. Are there any benefits from the balanced scorecard itself? (Follow up: Apart from the individual
measures?)
8. Do you have any (other) recommendations for improving the balanced scorecard?
The researchers asked essentially the same questions of administrators of the DBSC, but their interviews
tended to be more open and wide-ranging. To keep within the time available, the researchers usually did not
ask the administrators questions about specific DBSC measures (questions 3 and 4). Thus, distributor and
administrator interviews are not directly comparable on all questions. Because the administrator interviews are
less focused on the DBSC measures, this study uses them for background information. Unless otherwise
indicated, the analyses that follow refer to the distributor interviews only.
The interviews were conducted via conference calls conducted over a three-week period, with one
researcher asking initial and follow-up questions and the second researcher taking notes and capturing the
commentary on a laptop computer. After each interview, the two researchers conferred immediately to
complete abbreviated comments that might be difficult to decipher later. Interview files were copied intact
and archived in several locations.
Coding Interview Data
Coding Procedures
Two alternative coding procedures are (1) completely free coding unconstrained by prior theory or (2)
strict use of codes based on theoretical constructs. Both approaches have their adherents. However, it is
unusual for accounting researchers to enter the field free from preconceptions or prior theory. Miles and
Huberman [1994] argue that, when theory guides inquiry, it is efficient and realistic to begin with a conceptual
framework, and add “free” codes as the data suggest. The result is a hybrid approach that acknowledges
theoretical guidance (or bias) and permits empirical flexibility (or theory revision). The research used the
management control and organizational communication literatures surveyed earlier as a coding and analysis
guide, but modified the framework as the researchers delved into the data. Thus, the study contains elements
of both theory building and testing.
The computerized analysis method applies codes that reflect theoretical or empirical constructs to the
qualitative data – a sophisticated way to annotate and generalize interview transcripts. The researchers
predetermined codes for the interview data to reflect the interview questions – questions 1, 2, 4 to 8 and
15
twelve distributor-supplied measure questions for question 3.
21
The researchers also created codes that reflect
expected management control factors (e.g., Causality among measures), attributes of organizational
communication (e.g., Supports company culture), and impacts of the DBSC (e.g., Measure causes change). As
discussed earlier, some codes reflect additional concepts, revealed in the coding process (e.g., Weight of each
measure in determining overall DBSC scores). These codes were then applied to the interview data as
illustrated in Figure 2.
22
The study did not use the software specifically to search for or count specific words
or phrases. Choice of vocabulary is arbitrary, and words or phrases may not carry meaning outside of their
spoken context [Miles and Huberman, 1994]. Analysis, therefore, required reading, understanding, and coding
blocks of text in the context of each interview. This is the most subjective stage of the analysis, but in
addition to using both an interview protocol and a coding scheme the researchers took other steps to increase
the objectivity of the coding. Appendix 2 details these steps of the analysis.
FIGURE 2
Example of Coded Interview Text
The final list of codes, with frequencies by interview, is in table 2. Observe that for ease of later
exposition, the study collects related individual codes into large-pattern codes or “supercodes.” These
supercodes reflect ex ante theoretical constructs (e.g., Effective communication, Effective management control, Positive
outcomes) and are analogous to statistical factor models. The frequencies of the codes are an indication of
relative importance of each of these concepts, but frequency does not reflect intensity of feelings, nor does it
reflect relations among concepts. These attributes of the data may be discovered through additional analyses,
which are described next. One or a few talkative respondents did not dominate the coded comments, though
one distributor’s interview was briefer than the rest.
TABLE 2
Interview Codes and Frequencies by Interview
Relations Among Codes
Theoretically Supported Model
Figure 3 is a model of relations among employees’ perceptions of the management control and
organizational communication attributes of the DBSC that is based on the prior literature review and codes
applied during analysis. The arrows (? ) between the boxes reflect expectations about causal relations. The
research expected that both Effective management control and Effective communication in the design and use of the
DBSC would cause Strategy alignment, Effective motivation, and, ultimately, Positive outcomes. In contrast, the
16
research expected that “ineffective” factors could cause “negative” outcomes (in this case, only Conflict/tension
was observed qualitatively and coded
23
).
FIGURE 3
Theoretical Model of Management Control,
Communication, and the BSC
Observing Relations Among Codes
The relational-query capabilities of qualitative software, such as Atlas.ti, permit extensive exploration of
associations and possible causal hypotheses using coded interviewees’ perceptions of the DBSC. Assessing
the degree of relation among codes requires analysis of proximity and context of hypothesized relations (as in
figure 3). This is analogous to building a correlation table using a set of statistical measures, where the
frequency and nature of qualitative associations are building blocks of causality. The study assessed causality
by testing for multiple qualitative attributes of causal relations. Appendix 2 details the steps used to
operationalize this approach to analyzing and establishing evidence of causality within the study’s research
questions.
RESULTS
It is clear that distributors are aware of and understand the company’s diagnostic objective for the overall
DBSC. Representative comments that explain their awareness of the new measures and their links include:
“A lot of businesses tend to run with financial and market share measures, but those are pretty crude handles.
We have to get underneath with measures like quality and cycle time, and softer things like employee
development. That’s where the leverage of the business is. The others are results of what you’ve done” [3:154-
158].
24
“I think they are all linked. It’s hard to be a good manager in one area and not another” [9:118-119].
The first objective of this study is to find if the DBSC is perceived to possess the attributes of effective
organizational communication and management control devices. The second objective is to determine
whether these attributes can be causally related to goal alignment, motivation, and reported process or
decision changes.
As detailed in Appendix 2, where the research found specific, consistent, frequent patterns of association, the
researchers looked for further evidence of causality, based on coherence,
25
which is closely related to face
validity. This credible “story” of coherent causality is what distinguishes between findings of causality or mere
association. Table 3 and figure 4 summarize the results of an exhaustive audit of the coherence of specific,
consistent, and frequent associations.
These exhibits contain only those associations found to meet sufficient
causality criteria (complete data are in Appendix 2).
17
TABLE 3
Summary of Distributor-Response Supercode
Associations
FIGURE 4
Empirical Model of Distributors’ BSC
Perceptions
Overview of Data-Supported Model
Empirically associated quotations in the interview data, which are reflected by links in figure 4, support
the research questions in interesting ways.
26
Further analysis of all the paired codes in table 3 and figure 4
reveals answers to the study’s research questions and leads to recommendations for improving the
effectiveness of the DBSC.
Trimming the ex ante relations in figure 3, reflected in figure 4, has implications for understanding how
the BSC may cause management control and communication of strategy. On the “effective” side of figure 4,
Effective management control appears to cause Aligned with strategy and Effective motivation, which in turn appears to
cause Positive outcomes (e.g., perception of Improvement). These are consistent, strong associations between
specific factors that tell a coherent story, which the research interprets as evidence of causality. There is,
however, no consistent evidence of a direct link between Effective management control and Positive outcomes. In this
model, Effective management control affects Positive outcomes through Aligned with strategy and Effective motivation.
These data provide support for the “effective” form of the management control research question 2.
Surprisingly, there are no consistent links between perceptions of Effective communication and any other
DBSC model factor, which provides no support for the “effective” form of the communication research
question 1. In this case, the effective communication aspects of the BSC appear to be redundant to effective
management control.
On the “ineffective” side of the model, Ineffective management control, Ineffective communication, and Ineffective
motivation are associated or appear to be causally related. Furthermore, they appear to cause Conflict/Tension,
which provides support for “ineffective” forms of both research questions 1 and 2. This indicates that poorly
designed and implemented features of the BSC can do harm to the communication and control of strategy.
Several causal links involving Ineffective motivation and other factors were unexpected and will be discussed later.
The research now addresses each of the causal and associative links in the context of the two research
questions, referring to links in figure 4.
Question 1: Is the BSC an (in)effective communication device, creating strategic (non)alignment,
(in)effective motivation, and (negative)positive organizational outcomes?
18
Effective Communication ? Strategy Alignment / Effective Motivation ? Positive Outcomes
Unexpectedly, the study found no consistent evidence of specific relations between the attributes of
Effective communication and other DBSC-model factors. Overall this study does not support the “effective”
form of research question 1 that Effective communication is either associated with or causes Strategic alignment,
Effective motivation, or Positive outcomes.
27
Ineffective Communication ? Strategy Non-alignment / Ineffective Motivation ?
Conflict/Tension
Ineffective communication appeared to be largely independent of other “ineffective” DBSC factors. However,
though the study found little evidence of the impact of Effective communication, there was abundant evidence
that the DBSC administrators’ frequent use of One-way reporting is a direct cause of Conflict/tension (16 causal
links).
Unfortunately, the Conflict/tension appeared to be unproductive (i.e., no consistent links to Positive outcomes).
This may contribute to a climate of distrust and alienation that reduces the company’s and its distributors’
effectiveness. The company imposed DBSC measures and benchmarks without seeking input, and then used
the DBSC as a diagnostic control and an evaluation measure. Distributors felt ignored and trivialized because
of their non-involvement. However, they have little recourse because of the frequency of One-way reporting,
which was a common complaint. For example:
“No response (to my complaints), so we stand by our measure [of safety]. I’ve gotten no response to my
concerns, and I’m ‘PO’d’ at them on this subject. Any distributor who is green is a liar. No realistic way in hell
that that can happen. The nature of the work we do, we just can’t do this…. Do they have any idea what the
distributor environment is? They don’t care enough to reconcile issues, but the factor itself is important” [6:
116-123].
Partly as anticipated, the data provide support for the “ineffective” form of research question 1: Ineffective
communication, specifically One-way reporting, has largely negative consequences for acceptance, perceptions, and
reported uses of the DBSC.
Question 2: Is the BSC an (in)effective management control device, creating strategic
(non)alignment, (in)effective motivation, and (negative)positive organizational outcomes?
Effective Management Control ? Strategy Alignment ? Positive Outcomes.
As expected, BSC characteristics of Effective management control (Effective measurement, Comprehensive
performance, and Weight) appear to be causally linked with Strategy alignment (9 causal links with Key factors, 5 with
BSC important for business, and 9 with Traditional market share, respectively). Distributors perceive that having
reliable data leads to the ability to take actions that affect the new customer-relationship strategy (e.g., 17
causal links with Measure causes change), which might not have been feasible before the DBSC. For example, in
the use of customer satisfaction measures:
19
“We [now] give the work to an outside service. They call a couple of customers every day. We get input on a
list of questions. If the customer ends up not having a good experience, (now) we can get that info that day
and call the customer ourselves” [8: 119-122].
Because the DBSC measures Comprehensive performance, including the key financial and non-financial
measures, it is a reflection of overall success of managing critical factors (BSC important for business). Thus,
managers have a better feel for how they are managing the overall business for both current and future
results.
“The BSC is trying to give us a broader business set of measures of success than the more traditional financial
and market share. It wraps a set of things together that make sense for managing the business” [3: 5-7].
One of the company’s key strategic goals is to increase its traditional market share. The relatively heavy
Weight placed on this DBSC measure forces distributors (sometimes reluctantly) to align their goals with
improving the traditional market segment. They value their relationship with the company, and the DBSC
tells them what they must do to be a successful distributor, though they interpret this to mean they should
pursue improvements in traditional market share to the exclusion of other growth opportunities.
“If they care only about one-third of their business, then that’s good. It’s worth 28 points on the BSC. I’m red
and yellow there so there’s no hope to be green from all the other measures…. They are measuring only
(traditional) market penetration…. Balanced scorecard is certainly a misnomer” [2: 122-126].
By including Key factors, the DBSC causes distributors to diagnose problems and change their processes
and actions (Measure causes change) in significant ways. This leads to numerous recommendations to Modify
measures or BSC (11 causal links) – an example of potential interactive use of the DBSC [e.g., Simons, 2000, ch.
10]. Measuring the percentage of customers’ problems diagnosed within one hour, for example, also caused
most distributors to refocus their parts and service resources to building customer relationships, consistent
with the new strategy, rather than fully utilizing capacity – an example of diagnostic use of the DBSC
[Simons, 2000].
28
“This (measure) differentiates our businesses from our competition. It requires a complete change of ‘culture’
within the shop. Now we have to manage the service event instead of just scheduling work” [1:102-104].
In the past, distributors had favored large, complex service jobs that were relatively easy to schedule and
that could be counted on to occupy technicians and service space for blocks of time. Customers who had
simple service requirements were placed in the service queue in order of arrival, with no preferential
treatment, with the result that many began to take their simple jobs elsewhere and with the risk that they
would be lost as permanent customers. Distributors observed that:
“(One-hour diagnosis) requires a change in measurement and is creating a new mindset within the service
organization…We can’t schedule it; we have to provide the capacity and the process [1: 233-239]. [One-hour
diagnosis] tends to make us triage like a hospital and do the quick jobs first [2: 58-59]. I wasn’t an advocate at
the start, but now I am. It tells us how quickly we figure out what’s wrong so we can make an intelligent
20
statement to the customer, and so they can say ‘go ahead’ or not. We have been able to flow more jobs
through our shop by getting the quick, easy stuff through the shop. It lets us turn jobs quicker and avoids
embarrassing situations…It’s helping us, though it’s not easy to change the mentality, but it’s good” [6: 56-63].
Although there is no evidence of a direct link from Effective management control to Positive outcomes, the data
otherwise provide extensive support for the “effective” form of research question 2. Effective management control
using the DBSC appears to indirectly cause Positive outcomes through Strategic alignment.
Effective Management Control ? Effective Motivation ? Positive Outcomes
The DBSC’s motivational impacts were obvious overall and with respect to specific factors. Incentives
included both improved distributor business performance and successful contract renewals. The management
control design of the DBSC reflects the Causality of the DBSC description of the business, which causes
Meaningful rewards (8 causal links). Distributors believe that improving non-financial DBSC measures will result
in improved customer relationships and significant financial rewards.
“(Service utilization is) the most important number in the whole business [5: 102]. I gave the formula to our
guys that, if we bill our technicians out (on average) one more hour a day, we would put over $X million to our
(annual) bottom line. That’s the kind of magnitude were talking about” [4: 79-82].
29
The DBSC is successful as a motivational tool when it reflects relations between Strategic alignment and
distributor performance. For example, setting Appropriate benchmark targets for motivation goes hand-in-hand
with management control of Key factors (14 association links). Distributors do not object to tough, but
attainable goals.
“Good measurement. Don’t have a problem with that hurdle. Huge issue and can’t stress it enough. We have
about (xxx) labor hours. I can have one (accident) per year to be green. That’s a tight hurdle. It’s probably a
little tight right now” [4: 88-91].
Furthermore, setting attainable but tough DBSC goals (Appropriate benchmarks) motivates distributors to
change their decisions and processes (Measure causes change).
“80% of work is in 4-hour range in our service shop. Great number because the mentality in our shops
had been that we want that big overhaul, the long, lengthy jobs. But then, service efficiency suffers. We didn’t
turn many jobs and lost a lot of hours because there is a good chance of losing hours [on a large job]…Give
the company credit for the four-hour target. They thought about it; it’s probably the industry norm. Fo cusing
on this number has changed some of the culture or at least the thought process in the shop. We changed to the
little jobs and we can get the big jobs later. So our management has awakened to the fact that they can manage
their shops better using the one-hour diagnostic time and four-hour jobs to make their shops more efficient”
[4:47-59].
Relative performance evaluation allows each distributor to know his relative standing and what others are
doing, and thereby motivates distributors and gives them a tool for Improvement (7 causal links).
21
“(Gathering) the information and sharing it back to us, saying other distributors are X. I can look at it and see
how I am doing. Why am I different? I can use it as a lever to try to improve” [7: 123-125].
The data provide consistent evidence of causation and support the contention that perceptions of this
BSC’s effective management control characteristics lead to effective motivation, strategic alignment, and then
positive outcomes, in support of the second research question.
Ineffective Management Control ? Ineffective Motivation ? Conflict/Tension
This study found no consistent or frequent links between any of the elements of Strategy non-alignment and
other DBSC-model factors. However, Key factors that are poorly represented in the DBSC are associated with
numerous examples of other shortcomings. Notably, Inaccurate/subjective measures of Key factors (24 associations)
contribute to perceptions of Inappropriate benchmarks (8 causal links), which appear to cause widespread
Conflict/tension (9 causal links). This nexus of factors appears to be responsible for much of the Conflict/tension
caused by the use of the DBSC (17 out of 33 causal links), which reflected a lack of local autonomy and
participation in determining measures and targets. For example:
“(The measure is) a bunch of ‘hooey’ as far as keeping score, but for us running our business it’s an important
measure. What we do internally is what’s important, not if we get a ‘star’ on our shoe. This is one area if the
company wants to improve, we need to be a lot more consistent and define that criterion much more closely.
We routinely measured ourselves before the company did this. We gauge ourselves monthly on this one. We
ignore the BSC measure for our purposes, and use our own.” [1: 128-135]
“(Safety is a) hot button. The BSC uses a totally ludicrous measure, but the concept is great. I have written
four memos on this subject. I ran two plants before this. I have 100 technicians, and if those 100 have more
than one accident in a year, I’m in the red. Ridiculous!” [6: 109-111]
30
The data provide consistent evidence of causal paths connecting Ineffective management control, Ineffective
motivation, and Conflict/tension, which support the “ineffective” form of research question 2.
Other Findings
Strategy Alignment – Ineffective Management control or Ineffective Communication or
Ineffective Motivation – Positive Outcomes
The study found several unexpected causal relations and associations. Upon reflection, however, it is not
surprising that complaints about Inappropriate benchmarks are causes of recommendations to Modify measures or
BSC. Clearly, distributors, who have economic stakes in both their business’ success and contract renewal,
want relevant measures and attainable goals for DBSC Key factors. For example,
“Is the x% (benchmark percentage of technician hours used for training) appropriate? Hard to say. Probably
now, that would be a low number given (that)…the company will completely obsolete its own product line
soon. The need for training is much greater today than it has been in the past. Some companies will use
training dollars (rather than percentage of training hours). They are at like 5% (of revenues), which is much
22
higher than us. This raises a question in our minds. Do we do enough? We are concerned if we are reinvesting
enough in our employees.” [1: 183-190]
Additionally, the most numerous, consistent evidence (62 associations shown in dotted lines on the right
side of figure 4) shows that Key factors are associated with Inaccurate/subjective measures, Not understandable messages,
Inappropriate benchmarks, and Costly to measure. Distributors are frustrated when they perceive ineffective
implementation of DBSC factors that they believe are key to their business success and contract renewal.
Typical comments, for example involving the DBSC measure of training for salaried employees, include:
“(Training of salaried employees is) as critical (as for technicians) but harder to measure. We have to use some
guessing, because they are not paid hourly. Also, what’s training? Clearly going to a class during the workday,
but what about going after work? What percent of the total salaried hours is that?” [5: 155-160]
“For salaried people, it’s harder. We have to look at expense reports, and it’s a horrendous process. When you
bring this data collection problem to the company, they say we can’t do that either. They don’t even do it, and
they aren’t sure of the credibility of their number. From feedback from other distributors, they are just taking a
stab at it. We actually compile the numbers, but others are getting green scores for just a guess. We’re yellow
or red, and it’s a real number. The cost of the time isn’t worth it. But, it’s the right idea and the right thing to
do.” [1: 175-183]
Distributors’ frustrations were obvious when they realized that the DBSC was attempting to measure and
communicate important success factors, but that it was doing so ineffectively.
“We don’t grow much, so we need to find ways to expand. That’s all they pushed here (in contract renewal).
At another distributor, all they pushed on was customer satisfaction. Some areas if we (both) know we’re doing
a bad job and were red, they don’t seem to care…. Great tool but I’m not sure we are using it the way it
should be used” [8: 175-181]. “This is something we all should pay more attention to. We haven’t done as well
as we should have, but the goal means nothing to me because I’m so far away from it” [5: 122-125].
While the study did not anticipate these (and other similar) associations, their discovery provides ample
additional evidence of opportunities to improve the control and communication of strategy with the DBSC.
The study found relatively few instances of associations between Conflict/Tension and Positive Outcomes, which
may reflect both the relative newness of the DBSC and the top-down, one-way dialogue prevalent in the
company.
Summary of Results
The BSC is an innovative strategy communication and management control development. However, as
with all innovations, establishing its validity takes time, objective evidence, and careful analysis. There is
always the danger that promotional “hype” will promise more than a technique can deliver, which could lead
to disappointment, skepticism, and failure to recognize significant benefits, even if they are not as grand as
advertised. Kaplan and Norton [1996, 2000] bill the BSC as a complete, reliable strategic guide. It perhaps will
prove to be just that. However, there is limited objective evidence presented in support of this proposition.
23
For example, Ittner et al. [2000] do not find compelling evidence that a large bank’s BSC promoted increased
strategic awareness. More empirical evidence will be useful, because most of the BSC literature is either
normative prescription or uncritical reports of BSC “successes.” We believe this study provides a significant
contribution to the literature, of interest to both academics and managers.
The present study uses a method of analysis that moves management accounting field research in the
direction of more generalizability and internal validity than is apparent in most descriptive field research in the
area. While this qualitative approach can never achieve the external validity of statistical analysis of archival
data, perhaps it can aid researchers (and their critics) who seek to increase the objectivity and reliability of
field-study analysis.
Our findings are that, in at least one corporate setting, the BSC does present significant opportunities to
develop, communicate, and implement strategy – just as Kaplan and Norton aver. We find evidence that
managers respond positively to BSC measures by reorganizing their resources and activities, in some cases
dramatically, to improve their performance on those measures. More significantly, they believe that improving
their BSC performance is improving their business efficiency and profitability. Managers react favorably to
the BSC and heed its messages when:
• BSC elements are measured effectively, aligned with strategy, and reliable guides for changes,
modifications, and improvements
• The BSC is a comprehensive measure of performance that reflects the needs of effective management
• The BSC factors are seen to be causally linked to each other and tied to meaningful rewards
• BSC benchmarks are appropriate for evaluation and useful for guiding changes
• Relative BSC performance is a guide for improvement
However, problems of designing and implementing the BSC may be no different from those associated
with any major change in performance-measurement systems. The following factors were found to negatively
affect perceptions of the BSC and cause significant conflict and tension between the company and its
distributors.
• Measures are inaccurate or subjective
• Communication about the BSC is one-way (i.e., top-down and not participative)
• Benchmarks are inappropriate but used for evaluation
Though some of these adverse findings are associated with recommendations for improvement, most are
found to be causes of unproductive conflict and tension or a general atmosphere of ineffectiveness. For
example, the study found many examples of key factors that were ineffectively implemented in the company’s
BSC. Left uncorrected, these negatives could result in deteriorated relations, increasing “imbalance” in the
BSC as focus shifts to more objective, short-term financial measures [Ittner et al., 2000], and forfeiting the
24
communication and management control benefits of the BSC. For example, we can speculate that we did not
observe sufficient relations between Positive outcomes and Conflict/Tension because there is little dialogue (or
dialectical process) between the company and its distributors – we found only 6 associations. In this case it
appears that conflict simmers and rarely results in a positive outcome.
On the brighter side, the previous bullet points represent value-added and non-value-added BSC
activities. To successfully design, implement and use the BSC, organizations should enhance the former,
positive factors and eliminate or correct the latter, negative factors. It may be worth noting that the total
number of consistent links on the “ineffective” side of the model in figure 4 far outweighs those on the
“effective” side (154 to 59). Thus, the predominance of negative perceptions reflects many opportunities to
improve both communication and control of strategy. It seems likely that this ineffectiveness could be
resolved and the negative outcomes of unnecessary conflict and tension could be avoided at relatively low
cost (though it may require significant changes in attitudes). Possible solutions could be as simple as
improved dialogue between the company and its distributors regarding important but ineffectively measured
or poorly understood DBSC factors [e.g., Lindquist, 1995].
Limitations and Future Research
Even though many of this company’s managers and distributors apparently use the DBSC as a valid
representation of their business, we recognize that their reported perceptions may not be valid
representations of their actions. To our knowledge, however, there has been no rigorous, statistical test of the
claim that the BSC is, in fact, a causal model, which is the focus of our ongoing research.
Preliminary analysis of the statistical properties of the host company’s DBSC confirms many expected
causal relations and in particular shows the importance of modeling time lags between changes in investments
in internal processes, customer value, and financial performance. Consistent with distributors’ beliefs, we
have found that “upstream” changes may not result in tangible financial improvements for over a year.
“You will see very little change from quarter to quarter. Last quarter only one measure changed” [9: 121-124].
“I expect a three to five-year lag to see a significant impact of market penetration investments. I’m spending a
gazillion dollars on it, but returns will be in about five years. We’ll see some short-term returns soon, but the
big returns are five years down the road” [2: 148-151]. “My gut feeling is that it took two to three years to
reorganize and retrain, and four or five years later it started to pay off. I expect a quicker response now from
improving the fill rate and one-hour diagnosis” [6: 205-207]. I would think about half a year to a year for the
parts fill rate. Do well, and your reputation becomes known and you’ll see some effect in the financials. It’s a
matter of customer awareness that we’re doing something different here that will bring repeat business” [3:130-
133].
“People are very sensitive. They let us know if we are not living up to expectations. Some of our
customers are looking elsewhere to get parts because of stocking problems. Customers will react in a six-month
window” [1: 217-223]
25
Practical difficulties that are encountered in any statistical test of a BSC include:
• Changes in BSC measures and links as systems evolve to meet changing conditions
• Changes in organizations, markets, and personnel that may affect BSC structure and links
• Long lead times before effects are seen in lagging measures of performance
• No effects or negative results that may be attributed to “bad design” or “bad implementation” rather
than to the concept of the BSC as a causal model
• Desirable effects or positive results that may be caused by other, related (but omitted) factors, but are
attributed to the BSC
Making progress on controlling these factors offers opportunities for significant contributions to our
understanding of strategy communication, performance measurement, and evaluation characteristics of the
BSC.
Epilogue
Since the data collection for this project in mid-1999, the DBSC has undergone significant changes. The
company has added new measures and deleted some of the original ones; adjusted weightings; and
reconfigured categories. The company did not change benchmark targets of the retained measures. The most
notable changes came into effect at the end of 1999 when DBSC managers trimmed it from 30 to 20
measures. One major adjustment was continuing de-emphasis of the Learning and Growth category – it is
now eliminated from the DBSC, but the measures continue to be compiled on an annual basis. This
important area of performance was a casualty of unreliable measurement – but perhaps presents an
opportunity for improving the DBSC. Another change is enhancement of measures of new market share,
largely at the request of distributors facing significant growth opportunities in the new markets. In this case,
the company acceded to the wishes of distributors although the new market share measures were perceived to
be much less reliable than the traditional market share.
Company managers regard the DBSC project as an evolving process. Since the time the interviews were
conducted, the percent of distributors rated "green" has risen from 2 (6%) to 16 (52%), while the percent of
distributors rated "red" has fallen from 10 (32%) to only 1 (3%). The average overall BSC score has risen
from 67 points to 74 points (out of 100). Also and importantly, distributors on average have realized modest
but observable improvements in financial performance over the twelve quarters for which we have data. For
example and as shown below, the DBSC financial measure, distributor PBIT/Sales, has improved by 6.4
percent (average over all distributors) comparing the first four quarters of the scorecard process to the last
four quarters.
Eighteen of the 31 distributors experienced increases in PBIT/sales (average of 49.2 percent); 14 of them
experienced declines (average of -19.1 percent). The largest decline in PBIT/sales was -53.3 percent, and the
26
largest increase was +216.2 percent. From the first four to the last four quarters covered by this study,
distributors’ DBSC and PBIT/sales performance was distributed as follows.
PBIT/sales
increased (+)
PBIT/sales
decreased (-)
Total
Distributors whose DBSC score increased (+) 12 5 17
Distributors whose DBSC score decreased (-) 6 8 14
Total 18 13 31
Average percentage change in PBIT/Sales: First four
quarters to last four quarters
49.2%
(19.1%)
6.4 %
The actual distribution is marginally significantly different from a uniform distribution (p = 0.054), with
20 of the 31 distributors on the ++/-- diagonal, as one might expect. Both the company and its distributors
expect “upstream” improvements to take several years to flow through financial results, so the data available
for this study might not be sufficient to fully capture the effect of the DBSC. The changes in the DBSC and
increases in scorecard and financial performance have encouraged the company to continue managing with
the DBSC. Perhaps greater attention to the root causes of unproductive conflict surrounding the DBSC will
result in higher distributor acceptance, use, and performance.
27
REFERENCES
Abernethy, M. and P. Brownell. 1997. Management control systems in research and development
organizations: The role of accounting, behavior and personnel controls. Accounting, Organizations and
Society 22: 233-248.
Ahrens, T. and J. F. Dent. 1998. Accounting and organizations: Realizing the richness of field research. Journal
of Management Accounting Research 10: 1 – 39.
Amir, E. and B. Lev. 1996. Value-relevance of non-financial information: The wireless communications
industry. Journal of Accounting and Economics. 22(1-3): 3-30.
Amit, R. and P. Shoemaker. 1990. Strategic assets and organizational rent. Strategic Management Journal 14: 33-
46.
Antle, R. and J. S. Demski. 1988 The controllability principle in responsibility accounting. The Accounting
Review 63(4): 700-718.
Atkinson, A. A., R. Balakrishnan, P. Booth, J.M. Cote, T. Groot, T. Malmi, H. Roberts, E. Uliana, and A. Wu.
1997. New directions in management accounting research. Journal of Management Accounting Research 9:
70-108.
Atkinson A. A. and W. Shaffir. 1998. Standards for field research in management accounting. Journal of
Management Accounting Research 10: 41 – 68.
Banker, R. D. and S. Datar. 1989. Sensitivity, precision and linear aggregation of signals for performance
evaluation. Journal of Accounting Research 27: 21-39.
Banker, R. D. H. Chang, and S. K. Majumdar. 1993. Analyzing the underlying dimensions of firm
profitability. Managerial and Decision Economics 14(1): 25-36.
Banker, R. D., G. Potter, and R. G. Schroeder. 1995 An empirical analysis of manufacturing overhead cost
drivers,” Journal of Accounting and Economics 19(1): 115-137.
Banker, R. D., S. Lee, and G. Potter. 1996. A field study of the impact of a performance-based incentive
plan,” Journal of Accounting and Economics 21(2): 195-226.
Banker, R. D., G. Potter, and D. Srinivasan. 2000. An empirical investigation of an incentive plan that
includes non-financial performance measures. The Accounting Review 75: 65-92.
Barclay, M. J., D. K. Gode, and S. P. Kothari, 2000. The advantages of using earnings for compensation:
Matching delivered performance, University of Rochester working paper.
28
Barker, R. T. and M. R. Camarata. 1998. The role of communication in creating and maintaining a learning
organization: Preconditions, indicators and disciplines. Journal of Business Communication 35(4): 443-
467.
Barth, M. E. and M. F. McNichols. 1994. Estimation and market valuation of environmental liabilities relating
to superfund sites,” Journal of Accounting Research 32(Supplement): 177-219.
Baxter, J. A. and W. F. Chua. 1998. Doing field research: Practice and meta-theory in counterpoint. Journal of
Management Accounting Research 10: 69 – 87.
Becker, B. and M. Huselid. 1998. High performance work systems and firm performance: A synthesis of
research and managerial implications. Research in Personnel and Human Resources Management 16: 53-101.
Behn, B. K. and R. A. Riley. 1999. Using non-financial information to predict financial performance: The case
of the US airline industry. Journal of Accounting, Auditing & Finance 14(1): 29-56.
Berliner, C. and J.A. Brimson, eds. 1988. Cost Management for Today’s Advanced Manufacturing: The CAM-I
Conceptual Design. Boston, MA: Harvard Business School Press.
Brewer, P. C. 1998. National culture and activity-based costing systems: A note. Management Accounting Research
9(2): 241-260.
Coffey, A., B. Holbrook, and P. Atkinson. 1996. Qualitative data analysis: Technologies and representations.
Sociological Research Online <www.socresonline.org.uk/socresonline/1/1/4.html>
Daft, R. L. and A. Y. Lewin. 1993. Where are the theories for the “new” organizational forms? An editorial
essay. Organization Science 4: i-vi.
Dearden, J. 1969. The case against ROI control. Harvard Business Review. May-June: 124-135.
De Hass, M. and A. Kleingeld. 1999. Multilevel design of performance measurement systems: Enhancing
strategic dialogue throughout the organization. Management Accounting Research 10: 233-261.
Dillman, D. 1978. Mail and Telephone Surveys: The Total Design Method. New York: John Wiley & Sons.
Dixon, J. R., A. J. Nanni, and T. E. Vollman. 1990. The New Performance Challenge: Measuring Manufacturing for
World Class Competition. Homewood, IL: Dow Jones-Irwin.
Einhorn, H. J. and Hogarth, R. M. 1986. Judging probable cause. Psychological Bulletin, 99: 3-19.
Epstein, M. J. and J. Manzoni. 1997. The balanced scorecard and tableau de bord: Translating strategy into
action,” Management Accounting 79: 28-36.
29
Feltham, G. and J. Xie. 1994. Performance measure congruity and diversity in multi-task principal/agent
relations. The Accounting Review 69: 429-53.
Flamholtz, E.G.. 1979. The process of measurement in managerial accounting: A psycho-technical systems
perspective. Accounting, Organizations & Society 5: 31-42.
Foster, G. and M. Gupta. 1990. Manufacturing overhead cost driver analysis. Journal of Accounting and Economics
12(1-3): 309-37.
Foster, G. and M. Gupta. 1999. The customer profitability implication of customer satisfaction. Stanford
University and Washington University working paper.
Ghosh, D. and R.F. Lusch. 2000. Outcome effect, controllability and performance evaluation of managers:
Some field evidence from multi-outlet businesses. Accounting, Organizations and Society 25: 411-425.
Goodman, M. B. 1998. Corporate Communications for Executives. Albany, NY: SUNY Press.
Gordon, L. and V. Naranyan. 1984. Management accounting systems, perceived environmental uncertainty
and organization structure: An empirical investigation. Accounting, Organizations and Society : 33-47.
Govindarajan, V. 1984. Appropriateness of accounting data in performance evaluation: An empirical
examination of environmental uncertainty as an intervening variable. Accounting, Organizations and
Society : 125-135.
Govindarajan, V. and A. Gupta. 1985. Linking control systems to business unit strategy: Impact on
performance. Accounting, Organizations and Society 10: 51-66.
Grant, R. 1991. The resource-based theory of competitive advantage. California Management Review 33: 114-
135.
Holmstrom, B. 1979. Moral hazard and observability. Bell Journal of Economics 10: 74-91.
Hughes, K. E. 2000. The value relevance of non-financial measures of air pollution in the electric utility
industry. The Accounting Review 75(2): 209-228.
Huselid, M. 1995. The impact of human resource management practices on turnover, productivity, and
corporate financial performance. Academy of Management Journal 38: 635-672.
Huselid, M., S. Jackson, and R. Schuler. 1997. Technical and strategic human resource management
effectiveness as determinants of firm performance. Academy of Management Journal 40: 171-188.
Ittner, C.D. and D.F. Larcker. 1997. Quality strategy, strategic control systems, and organizational
performance. Accounting Organizations and Society 22(3/4): 293-314.
30
-----------. 1998a. Are non-financial measures leading indicators of financial performance? An analysis of
customer satisfaction. Journal of Accounting Research 26(supplement): 1-34.
-----------. 1998b. Innovations in performance measurement: Trends and research implications. Journal of
Management Accounting Research 10: 205-238.
Ittner, C. D., D. F. Larcker, and M. W. Meyer. 2000. The use of subjectivity in multi-criteria bonus plans. The
Wharton School working paper.
Johnson, H. T. 1992. Relevance Regained: From Top-Down Control to Bottom-Up Empowerment. New York, NY: The
Free Press.
Johnson, H. T. and R. S. Kaplan. 1987. Relevance Lost: The Rise and Fall of Management Accounting. Boston, MA:
Harvard Business School Press.
Kaplan, R. S. and D. P. Norton. 1992. The balanced scorecard – Measures that drive performance. Harvard
Business Review January – February: 71-79.
-----------. 1993. Putting the balanced scorecard to work. Harvard Business Review September – October: 143-
142.
-----------. 1996a. Using the balanced scorecard as a strategic management system. Harvard Business Review
January – February: 75-85.
-----------. 1996b. The Balanced Scorecard. Boston, MA: Harvard Business School Press.
-----------. 1996c. Linking the balanced scorecard to strategy. California Management Review, Fall: 53-79.
-----------. 2000. The Strategy-Focused Organization. Boston, MA: Harvard Business School Press.
Keegan, D. P., R. G. Eiler, and C. R. Jones. 1989. Are your performance measures obsolete? Management
Accounting :45-50.
Kotter, J. P. 1995. Why transformation efforts fail. Harvard Business Review, Mar-Apr: 61.
Lee, R. and N. Fielding. 1996. Qualitative data analysis: Representations of a Technology: A comment on
Coffey, Holbrook and Atkinson. Sociological Research Online . WWW page at URL
www.socresonline.org.uk/socresonline/1/4/lf.html
Lillis, A. M. 1999. A framework for the analysis of interview data from multiple field research sites Accounting
& Finance 39: 79-105.
31
Lindquist, T. 1995. Fairness as an Antecedent to Participative Budgeting: Examining the Effects of
Distributive Justice, Procedural Justice and Referent Cognitions on Satisfaction and Performance.
Management Accounting Research, 7: 122-147.
Lipe, M. G. and S. Salterio. 2000. The balanced scorecard: Judgmental effects of information organization
and diversity. The Accounting Review.
Locke, E. A. and G. P. Latham. 1990. A Theory of Goal Setting and Task Performance. Englewood Cliffs, NJ:
Prentice Hall.
Luft, J. and M. D. Shields. 1999. Accounting classification of expenditures on intangibles: Cognitive causes
of managerial myopia. Working paper, Michigan State University.
Lynch, R. L. and K. F. Cross. 1995. Measure Up! Yardsticks for Continuous Improvement. Cambridge, MA:
Blackwell Business.
Malina, M.A. 2001. Management control and the balanced scorecard: An empirical test of causal relations.
University of Melbourne working paper.
McKenzie, F. C. and M. D. Schilling. 1998. Avoiding performance measurement traps: Ensuring effective
incentive designs and implementation. Compensation and Benefits Review 30(4): 57-65.
Merchant, K. A. 1989. Rewarding Results: Motivating Profit Center Managers. Boston, MA: Harvard Business
School Press.
Miles, M. B. and A. M. Huberman. 1994. Qualitative Data Analysis. Thousand Oaks, CA: SAGE Publications.
Nanni, A. J., J. G. Miller, and T. E. Vollmann.1988. What shall we account for? Management Accounting v69n7,
(Jan): 42-48.
Norreklit, H. 2000. The balance on the balanced scorecard – A critical analysis of some of its assumptions.
Management Accounting Research 11: 65-88.
Perera, S., G. Harrison, and M. Poole. 1997. Customer-focused manufacturing strategy and the use of
operations-based non-financial performance measures: A research note,” Accounting Organizations and
Society 22(6): 557-572.
Rappaport, A. 1999. New thinking on how to link executive pay to performance. Harvard Business Review. Mar-
Apr: 91-101.
32
Schulze, W. 1992. The two schools of thought in resource-based theory: Definitions and implications for
research. Paper presented at the annual meeting of the Academy of Management, Las Vegas,
Nevada.
Silk, S. 1998. Automating the balanced scorecard. Management Accounting 79(11): 38-42.
Simons, R. 2000. Performance Measurement & Control Systems for Implementing Strategy. Upper Saddle River, NJ:
Prentice Hall.
Trochim, W.M. 2000. The Research Methods Knowledge Base, 2nd Edition. Internet WWW page, at URL:
<http://trochim.human.cornell.edu/kb/index.htm>.
Tucker, M. L., G. D. Meyer, and J. W. Westerman. 1996. Organizational communication: Development of
internal strategic competitive advantage. Journal of Business Communication 33(1): 51-69.
Ulrich, D. and D. Lake. 1990. Organizational Capability: Competing from the Inside Out. New York, NY: Wiley.
Watson, D. J. and J. V. Baumler. 1975. Transfer pricing: A behavioral context. The Accounting Review 50: 466-
74.
West, G. P. and G. D. Meyer. 1997. Communicated knowledge as a learning foundation. The International
Journal of Organizational Analysis, 5(1): 25-58.
Yin, R. K. 1994. Case Study Research: Design and Methods, 2
nd
ed. Applied Social Research Methods Series,
Volume 5. Thousand Oaks, CA: Sage Publications.
33
TABLE 1
DBSC Measures and Approximate Weights
Traditional BSC Categories Distributor BSC Measures (Company category) Weights
Learning and growth Employee skill inventory and personal development plans (HC)… 1%
Industry involvement (HC)……………………………………….. 1%
Training (HC) ……………………………………………………. 2% 4%
Efficient internal processes Customer orders, first-time fill rate (CA)………………………… 3%
Customer service, problems diagnosed in 1 hour (CA)…………… 5%
Customer service, problems solved in 6 hours (CA)…………….. 5%
Management excellence awards (CA) ……………………………. 3%
Adoption of best practices (CA) ………………………………… 1%
Inventory turnover, (PG) ……………………………………….. 4%
Days sales outstanding (PG) …………………………………….. 2%
Service hours utilization (PG) …………………………………… 2%
Safety (CC) ……………………………………………………… 2%
Warranties (Other) ………………………………………………. 8%
Building condition (Other) ……………………………………….. 3%
Miscellaneous (Other) …………………………………………… 3% 41%
Customer value Customer satisfaction (CA) ……………………………………….. 4%
Traditional market share – 1 (easily tracked) (CA)………………….
28%
New market share – 2 (no measure yet available) (CA)……………. 6%
Environmental assessment and remediation (CC)…………………. 2% 40%
Financial success PBIT, % of sales (PG) …………………………………………… 4%
Cash flow from operations, % of sales (PG)……………………… 2%
Sales growth (PG) …………………………………………………
9% 15%
100%
Company BSC categories:
HC = Investments in human capital
CA = Competitive advantage
PG = Profitability and growth
CC = Corporate citizenship
34
FIGURE 1
Representative DBSC Ratings and Scores
Competitive Advantage
Customer satisfaction
Traditional market share – 1
New market share – 2
Customer orders, first-time fill rate*
Customer service, problems diagnosed in 1 hour *
Customer service, 4-hour problems solved in 6 hours*
Management excellence awards
Adoption of best practices*
Total Competitive Advantage Rating
Profitability and Growth
PBIT, % of sales
Cash flow from operations, % of sales
Inventory turnover
Days sales outstanding
Service hours utilization*
Sales growth
Total Profitability and Growth Rating
Corporate Citizenship
Environmental assessment and remediation
Safety*
Total Corporate Citizenship Rating
Investments in Human Capital
Employee skill inventory and personal development plans*
Industry involvement*
Training*
Total Human Capital Rating
Other
Warranties
Building condition
Miscellaneous
Total Other
Total – Overall Rating
Total – Overall Score
Weights, % 4 28 6 3 5 5 3 1 55 4 2 4 2 2 9 23 2 2 4 1 1 2 4 8 3 3 14 100
100
Distributor
1 Y Y Y Y Y R R Y G R Y G Y G G Y Y Y R Y Y Y Y Y R Y R 64
2 Y G G Y Y G G G G Y Y Y Y G G G Y G G G G G G G G G G 84
3 R Y Y Y Y G R Y Y R Y Y Y Y Y Y Y Y R Y Y Y G Y G Y Y 67
4 R Y Y Y R G G Y Y Y R R R G Y G R Y R G Y Y R Y R R R 61
5 R G Y Y Y G R Y G G R R Y G Y G R Y G G Y G G Y G G G 75
6 Y Y Y Y Y G G Y G G Y Y Y G G R R R Y G Y Y Y Y Y Y Y 70
* Supplied by distributors to the company (22% weighting). Other items prepared by the company from financial and statistical reports (78% weighting).
35
FIGURE 2
Example of Coded Interview Text
Note: In this example, the highlighted quotation in the left-hand window corresponds to the code meaningful reward/penalty, which is outlined in a dotted box
in the right-hand window. Clicking on each of the codes on the right would highlight corresponding quotations on the left. Note also that the code
causality among measures encloses the code meaningful reward/penalty. The bracket to the left of each code shows its location.
36
TABLE 2
Interview Codes and Frequencies by Interview
Distributors
Interview
1 2 3 4 5 6 7 8 9
Overall BSC rating at time of interview
Y Y R Y Y G Y R R
Overall BSC score at time of interview
72 67 62 74 67 84 70 61 60 Totals
sc-Interview protocol* 17 18 17 17 17 17 17 16 17 152
?benefits of BSC 1 1 1 1 1 1 1 1 1 9
?best practices 1 1 1 1 1 1 1 1 1 9
?company’s measures 1 1 1 1 1 1 1 1 1 9
?definition of BSC 1 1 1 1 1 1 1 1 1 9
?diagnose-1-hour 1 1 1 1 1 1 1 1 1 9
?first-pass fill rate 1 1 1 1 1 1 1 1 1 9
?industry involvement 1 2 1 1 1 0 1 1 1 9
?objective/purpose of BSC 1 1 1 1 1 1 1 1 1 9
?other recommendations 1 1 1 1 1 1 1 1 1 9
?personnel reviews 1 1 1 1 1 2 1 0 1 9
?relations among measures 1 1 1 1 1 1 1 1 1 9
?safety 1 1 1 1 1 1 1 1 1 9
?service utilization 1 1 1 1 1 1 1 1 1 9
?solve-4-hour 1 1 1 1 1 1 1 1 1 9
?training-salary 1 1 1 1 1 1 1 1 1 9
?training-technician 1 1 1 1 1 1 1 1 1 9
?useful measure 1 1 1 1 1 1 1 1 1 9
sc- Strategy alignment * 23 20 20 12 22 20 11 12 12 152
Key factors 12 11 13 10 13 16 8 10 6 99
BSC important for business 3 3 4 2 4 3 2 2 2 25
Imbalanced market share 3 4 1 0 3 0 1 0 4 16
Support company strategy 5 2 2 0 2 1 0 0 0 12
Strategy non-alignment 2 2 1 2 0 0 0 1 5 13
sc-Effective communication* 2 1 6 6 4 6 3 3 2 33
Routine 1 0 3 4 2 1 2 0 1 14
Support company culture 1 0 0 0 1 1 0 2 0 5
Trustworthy 0 0 0 0 0 0 0 0 0 0
Two-way dialogue 0 0 3 2 1 3 1 1 1 12
Understandable 0 1 0 0 0 1 0 0 0 2
sc-Ineffective communication* 8 10 10 6 10 13 13 12 10 92
Not routine 0 0 1 1 0 1 0 0 1 4
Not support company culture 0 0 0 0 0 0 0 0 0 0
Not trustworthy 0 3 1 1 4 1 0 1 1 12
Not understandable 4 4 6 3 4 1 0 1 5 28
One-way reporting 4 3 2 1 2 10 13 10 3 48
* The “sc” prefix refers to a “supercode,” which collects indented codes listed below the supercode.
37
Table 2 (continued)
Distributors
Interview
1 2 3 4 5 6 7 8 9
Most recent overall BSC rating
Y Y R Y Y G Y R R
Most recent overall BSC score
72 67 62 74 67 84 70 61 60 Totals
sc-Effective management
control*
12 13 12 9 18 9 7 4 6 90
Causality among measures 3 1 5 4 6 4 3 1 2 29
Comprehensive performance 4 1 2 1 3 1 1 1 1 15
Effective measurement 4 4 3 3 4 2 2 2 0 24
Time lag 1 1 1 1 0 1 0 0 1 6
Weight 0 6 2 0 5 1 1 0 2 17
sc-Ineffective mgt control* 11 7 7 4 7 4 3 2 6 51
Inaccurate/subjective measure 9 6 5 3 7 4 3 2 5 44
Limited scope of measure 2 1 2 1 0 0 0 0 1 7
sc-Effective motivation * 4 11 15 10 8 11 13 9 5 86
Absolute performance 0 0 0 0 1 1 1 1 0 4
Appropriate benchmark 0 2 4 5 2 2 5 2 2 24
Meaningful reward/penalty 0 3 1 1 3 4 1 2 1 16
Motivate distributors 1 1 8 0 0 1 0 2 0 13
Objective performance 0 2 0 1 0 2 0 2 1 8
Relative performance 3 3 2 3 2 1 6 0 1 21
sc-Ineffective motivation* 4 11 15 10 8 11 13 9 5 86
Costly to measure 2 3 2 0 0 2 5 3 1 18
Inappropriate benchmark 6 1 0 2 6 4 4 6 4 33
Ltd controllability by distrib. 3 1 0 4 0 5 1 2 3 19
Subjective performance 0 0 1 0 1 0 0 4 1 7
sc-Positive outcomes* 12 5 14 11 4 10 10 10 11 87
Improvement 3 2 3 2 0 1 3 0 1 15
Measure causes change 5 2 6 6 1 4 1 6 3 34
Modify measure or BSC 4 1 5 3 3 5 6 4 7 38
Conflict/tension 6 2 2 1 2 6 3 4 2 28
* The “sc-“ prefix refers to a “supercode,” which collects indented codes listed below the supercode.
38
TABLE 3
Summary of Verified Supercode Causal Relations and Associations
First Supercode Second Supercode Causal Relations
Associations
Effective Mgt Control……
Strategy Alignment…………
23
Effective Mgt Control……
Effective Motivation……….
8
Strategy Alignment……….
Positive Outcomes…………
28
Strategy Alignment……….
Ineffective Mgt Control…..
.………………….
24
Strategy Alignment……….
Effective Motivation …….
…………………..
14
Strategy Alignment……….
Ineffective Communication
…………………..
11
Strategy Alignment……….
Ineffective Motivation…….
…………………..
27
Effective Motivation………
Positive Outcomes ……….
14
Ineffective Mgt Control…. Conflict/Tension………….
8
Ineffective Mgt Control…...
Ineffective Motivation……
8
Ineffective Mgt Control…...
Ineffective Communication
………………
…..
16
Ineffective Communication
Conflict/Tension…………
16
Ineffective Motivation…….
Positive Outcomes……….
7
Ineffective Motivation…….
Conflict/Tension…………
9
39
Research Question 2:
Effective Management Control
(((("causality among measures" |
"comprehensive performance") |
"Effective measurement") |
"time lag") | "weight")
Strategy Alignment
((("traditional market
share" | "support
company strategy") | "BSC
important for business") |
"key factors")
Positive Outcomes
(("improvement" |
"measure causes change")
| "modify measure or
BSC")
Research Question 1:
Effective Communication
(((("routine" | "support company
culture") | "two-way dialogue") |
"trustworthy") |
"understandable, clear
presentation")
Effective Motivation
((((("objective
performance" | "absolute
performance") | "relative
performance") |
"appropriate benchmark")
| "motivate distributors")
| "meaningful reward/
penalty")
Strategy
Non-Alignment*
Research Question 2:
Ineffective Management Control
("inaccurate/subjective measure"
| "limited scope of measure")
Research Question 1:
Ineffective Communication
(((("not routine" | "not support
company culture") | "one-way
reporting") | "not trustworthy")
| "not understandable")
Ineffective Motivation
((("subjective
performance" |
"inappropriate
benchmark") | "limited
controllability by
distributor") | "costly to
measure")
Conflict/Tension*
Figure 3
Theoretical Model of Communication, Management Control, and the BSC
* Single-item codes
40
Causality
Appropriate
Benchmark
RQ2: Effective
Management
Control
Comprehensive
Performance
Effective
Measurement
Weight
Strategy Alignment
Key factors
BSC
Important
for Business
Traditional
Market Share
Relative Performance
Meaningful Reward
Measure Causes
Change
Effective
Motivation
Positive
Outcomes
Modify Measure or
BSC
Inappropriate
Benchmark
RQ2: Ineffective
Management Control
RQ1: Ineffective
Communication
Ineffective
Motivation
Inaccurate/Subjective
Measure
Not
Understandable
Costly to Measure
24
11
17
10
16
8
Improvement
Data-Supported Model of Distributors' BSC Perceptions
14
Figure 4
Conflict/
Tension
One-way
Reporting
9
RQ1: Effective
Communication
Strategy Non-Alignment
Legend
Causality
("hits")Association
("hits")
9
5
9
16
8
17
11
7
7
7
18
41
APPENDIX 1: CODES AND DEFINITIONS
?benefits of BSC Question refers to beneficial effects of BSC
?best practices Question refers to submitting a best practice for others to consider adopting
?company's measures Question refers to BSC elements measured by company, not the distributor
?definition of BSC Question about perception of what the BSC is supposed to be
?diagnose-1-hour Question about part of program to accurately diagnose service problem
within one hour
?first-pass fill rate Question about measure of ability to fill customer request for parts from
available inventory - 100% is tops
?industry involvement Question about whether distributors are "networking" with customers on a
professional basis
?objective/purpose of BSC Question about the objective of the BSC, what the BSC is supposed to
achieve
?other recommendations Question about miscellaneous recommendations for BSC improvement
?personnel reviews Question about proportion of employees who have annual performance
plans and reviews
?relations among measures Question whether BSC measures are related
?safety Question about safety, one of the BSC measures, relates number of time-
loss incidents
?service utilization Question about utilization of service capacity - hours of productive time
?solve-4-hour Question about proportion of standard 4-hour repair jobs completed within
6 hours
?training-salary Question about training and skill achievement of salaried, non-technicians
?training-technicians Question about training and skill achievement of technicians
?useful measure Question about whether BSC and/or measures are useful for managing the
business
Absolute performance BSC or individual measure for comparison of performance against a
standard of performance, not necessarily relative to other distributors
Appropriate benchmark Benchmark is appropriate to achieve red, yellow, or green ratings;
challenging but attainable
BSC important for business The BSC as a whole is important for managing the business; reference to
more than a single measure
Causality among measures BSC design reflects cause and effect relations among measures
Comprehensive performance The BSC is meant to or does provide an overall (financial and non-financial)
measure of performance
Conflict/tension Evidence of conflict or tension caused by the BSC, its elements, or its use
Costly to measure Measure is costly/difficult/time consuming to measure or maintain
Cross-product subsidy Evidence that costing results in mis-allocations of cost
Effective measurement Measure is hard, verifiable, valid - measures what it says it is measuring
Imbalanced market share Emphasis on traditional market share, not new markets
Improvement BSC can be/is tool to improve business; evidence or strong belief of
improvement of performance
Inaccurate/subjective
measure
Measure does not reflect underlying key activity, not consistently reported
over time and across distributors, soft measure, system not capturing
measure correctly
Inappropriate benchmark Not an appropriate benchmark for achieving red, yellow, or green ratings;
not attainable or too easy
42
Inappropriate factor Measure is not important for measuring performance of the business, from
either distributor or company perspective
Key factors Individual measure is important for managing the measure from either
distributor or company perspective; good, important, helpful to measure
Limited controllability by
distributor
Measure is largely outside the control of the distributor
Limited scope of measure Measure does not reflect full extent of important distributor activity; focus
too narrow
Meaningful reward/penalty Materially affects distributor profits, compensation, or the 3-year review for
renewal of distributor license
Measure causes change Use of measure causes change in distributor action or mindset
Modify measure or BSC Argument that the company should revise the BSC measure to be more
accurate or appropriate; specific recommendation
Motivate distributors BSC motivates distributors to improve measures and/or performance
Not aligned with strategy Measure is not important for measuring performance of the business, from
either distributor or company perspective
Not routine An ad hoc measure, difficult to predict when needed or available
Not support company culture Measure works against the company's culture, values, or beliefs
Not trustworthy Measure cannot be trusted because of manipulation, bias, or outright
cheating that cannot be detected
Not understandable Distributors do not understand what the measure is or what it is trying to
represent; measure poorly defined
Objective performance BSC rating or score is or is supposed to be an objective measure of
performance
One-way reporting Communication is top-down, dictated policy or prescription; mandatory
reporting, no feedback, no dialogue among distributors, no input to
company
Relative performance Measure or BSC is used to evaluate distributors relative to each other, apart
from against a standard of performance
Routine Measure is regularly available or used; easy to predict when needed or
available
Subjective performance Actual performance evaluations are wholly or in part subjective
Support company culture Measure reinforces company culture, values, beliefs
Support company strategy Measure identifies or reinforces company strategic plans or initiatives
Time lag Evidence that downstream effects are lagged from upstream activities or
investments; amount of time before a change in one measure is reflected in
another downstream measure
Trustworthy Measure is trustworthy, reliable and free from manipulation, bias or
undetectable cheating
Two-way dialogue Communication is open dialogue - sharing of views and ideas among
company and distributors, includes feedback
Understandable BSC presents measures and performance in clear, understandable format
Weight Weight reflects importance to company unless the measure cannot be made
objectively
43
APPENDIX 2: STEPS TO ASSURE CODING RELIABILITY AND ESTABLISH CAUSALITY
Insuring Coding Reliability
After agreeing upon the predetermined coding scheme, each of the two researchers coded the first interview
using the software tool. After coding the first interview, the researchers met, computer files side by side, to compare
coding, resolve differences, and agree on a refined set of codes (reduced from 70 to 54). The researchers then coded
the remaining interviews, and mutually resolved any disagreements. In some instances, the resolution was to revise
the name or definition of a code. A small number of preset codes were not used and are not reported.
To test coding reliability, several weeks later the researchers jointly recoded three randomly selected interviews.
31
The researchers then noted the number of agreements and disagreements between the first and second codings. The
software allows the researcher to code any portion of text – a single word, phrase, sentence, paragraph, and so on.
Therefore, the researchers did not count minor differences in boundaries of text blocks as disagreements; rather, a
“disagreement” was a different code (or no code) applied to roughly the same block of text. An “agreement” was
using the same code for approximately the same block of text. Coding agreement for this test averaged 80.3 percent
[agreements/(agreements + disagreements)], ranging from 69 percent to 87 percent across the three interviews.
This level of coding reliability is within norms of 80 to 90 percent coding reliability [Miles and Huberman, 1994, p.
64].
32
Finding Associations Among Codes
The qualitative software (Atlas.ti) easily enables queries of proximity relations or associations among coded
quotations listed below. Examples in parentheses refer to codes illustrated in figure 2.
• Coded quotations of one type enclose coded quotations of another type (inaccurate/subjective measure, lines
0075-0079, encloses not understandable, lines 0075-0077)
• Coded quotations of one type are enclosed by coded quotations of another type (not understandable, lines
0075-0077, enclosed by inaccurate/subjective measure, lines 0075-0079)
• Coded quotations of one type overlap coded quotations of another type (causality among measures, lines
0079-0082, overlaps meaningful reward/penalty, lines 0080-0082)
• Coded quotations of one type are overlapped by coded quotations of another type (meaningful
reward/penalty, lines 0080-0082, overlapped by causality among measures, lines 0079-0082)
• Coded quotations of one type precede coded quotations of another type by no more than one line
33
(not
understandable, lines 0075-0077, precedes not trustworthy, line 0077)
• Coded quotations of one type follow coded quotations of another type by no more than one line (not
trustworthy, line 0077, follows not understandable, lines 0075-0077)
The full results of this analysis are in table 4, and in summary form in table 3.
44
TABLE 4
Distributor-Response Supercode Associations
Establishing Causality Among Codes
Close proximity or association of types of quotations might indicate causality (similar to statistical correlation), but
analysis of the context of these measures of proximity is necessary. Miles and Huberman [1994, p. 146-7]
demonstrate that qualitative analysis uses the same rules of causality as statistical analysis. Investigating the context
and meaning of associations in qualitative data may reveal causality (for example, between Effective management control
– EMC – and Strategy alignment – SA) by any of the following observations (the more the better), in rough order of
applicability to this study:
34
• Specificity (a particular link is shown between EMC and SA)
• Consistency (EMC is found with SA in different places)
• Strength of association (much more SA with EMC than with other possible causes)
• Coherence (EMC-SA relationship fits with what else is known about EMC and SA)
• Plausibility (a known mechanism exists to link EMC and SA)
• Temporality (EMC before SA, not the reverse)
• Behavioral gradient (if more EMC, then more SA)
• Analogy (EMC and SA resemble the well-established pattern in other relations)
• Experiment (change EMC, observe what happens to SA)
Using supercode-level queries reduced the number of specific associations dramatically. Queries will find every
association of the elements of supercodes, not all of which may be evidence of causality. To focus the investigation
on consistent links, the research identified all supercode links with a total of 27 or more “hits” or observed
associations,
35
and looked for concentrated evidence of causality between individual or sub-coded comments. The
number of paired sub-code relations rarely exceeded 10 hits. Therefore, to avoid omitting some meaningful sub-
code relations, the audit was expanded to all supercode relations with 10 or more hits. In some cases, the total
number of hits linking two supercodes was widely diffused across their elements, with no consistent patterns at the
individual code level. The research did not investigate these diffused associations further. That is, the research
conservatively treated consistent, strong (i.e., frequent) relations of specific factors as necessary for establishing causation
in this study. To qualify as evidence of causality, each of these associations also must demonstrate coherence or be
consistent with an explanation of causality.
For example, nearly all cases of consistent, frequent, and specific associations found by the “encloses,”
“enclosed by,” “overlaps,” and “overlapped by” operators also presented coherent stories of causality. For example
in figure 2, one can infer causality from the observed relation between causality among measures (lines 0079-0082) and
45
meaningful reward/penalty (lines 0080-0082) – because the DBSC reflects causal relations among measures (part of
Effective management control), distributors expect to achieve meaningful rewards by using it (part of Effective motivation).
However, some of the associations found by the “precedes” or “follows” operators were only coincidentally
proximate. That is, these were associations for which the research could not develop a coherent causal story linking
the two codes and were deleted from the counts of evidence of causality.
46
TABLE 4
Analysis of Distributor-Response Supercode Proximity and Association
First Code Second Code Enclosed byEncloses
Overlapped by
Overlaps
Follows*
Precedes*
Total Associations
Effective Mgt Control Aligned with Strategy 12 7 2 0 11 6 38
Effective Mgt Control Conflict/Tension 2 0 0 0 1 2 5
Effective Mgt Control Effective Communication
1 1 0 0 1 2 5
Effective Mgt Control Effective Motivation 2 7 1 1 11 12 34
Effective Mgt Control Ineffective Communication
0 0 1 0 8 4 13
Effective Mgt Control Ineffective Mgt control 0 0 0 0 5 6 11
Effective Mgt Control Ineffective Motivation 0 0 0 0 4 5 9
Effective Mgt Control Strategy Non-Alignment 0 0 0 0 0 1 1
Effective Mgt Control Positive Outcomes 3 3 1 0 6 5 18
Ineffective Mgt Control Aligned with Strategy 2 2 0 0 18 7 29
Ineffective Mgt Control Conflict/Tension 3 1 0 1 2 3 10
Ineffective Mgt Control Effective Communication
0 0 0 0 0 1 1
Ineffective Mgt Control Effective Mgt Control 0 0 0 0 5 5 10
Ineffective Mgt Control Effective Motivation 0 1 0 0 8 5 14
Ineffective Mgt Control Ineffective Communication
3 9 2 1 8 8 31
Ineffective Mgt Control Ineffective Motivation 4 0 0 2 7 5 18
Ineffective Mgt Control Strategy Non-Alignment 0 1 0 0 0 0 1
Ineffective Mgt Control Positive Outcomes 0 3 0 0 1 12 16
* Follows or precedes the first code by one line. See footnote 18.
continued
47
Table 4 (continued)
First Code Second Code Enclosed by
Encloses
Overlapped by
Overlaps
Follows*
Precedes*
Total Associations
Effective Communication
Strategy Alignment 4 2 0 0 6 1 13
Effective Communication
Conflict/Tension 0 0 1 1 0 3 5
Effective Communication
Effective Mgt Control 1 1 0 0 2 1 5
Effective Communication
Effective Motivation 4 0 0 0 0 3 7
Effective Communication
Ineffective Communication
1 0 1 0 5 4 11
Effective Communication
Ineffective Mgt Control 0 0 0 0 1 0 1
Effective Communication
Ineffective Motivation 1 0 0 0 2 4 7
Effective Communication
Strategy Non-Alignment 0 0 0 0 1 1 2
Effective Communication
Positive Outcomes 2 0 0 0 4 4 10
Ineffective Communication
Strategy Alignment 3 2 0 1 14 10 30
Ineffective Communication
Conflict/Tension 8 3 1 1 2 4 19
Ineffective Communication
Effective Communication
0 1 0 1 5 5 12
Ineffective Communication
Effective Mgt Control 0 0 0 1 4 8 13
Ineffective Communication
Effective Motivation 3 2 0 0 8 11 24
Ineffective Communication
Ineffective Mgt Control 13 4 1 2 8 8 36
Ineffective Communication
Ineffective Motivation 7 2 1 2 7 4 23
Ineffective Communication
Strategy Non-Alignment 0 1 0 0 3 2 6
Ineffective Communication
Positive Outcomes 3 3 0 0 5 9 20
Strategy Alignment Conflict/Tension 4 1 0 0 5 3 13
Strategy Alignment Effective Communication
2 4 0 0 1 6 13
Strategy Alignment Effective Mgt Control 7 12 0 2 6 11 38
Strategy Alignment Effective Motivation 6 7 1 1 5 11 31
Strategy Alignment Ineffective Communication
2 1 1 0 10 14 28
Strategy Alignment Ineffective Mgt Control 2 1 0 0 8 19 30
Strategy Alignment Ineffective Motivation 1 1 2 1 10 22 37
Strategy Alignment Strategy Non-Alignment 0 1 0 0 2 1 4
Strategy Alignment Positive Outcomes 11 8 1 0 12 21 53
continued
48
Table 4 (continued)
First Code Second Code Enclosed by
Encloses
Overlapped by
Overlaps
Follows*
Precedes*
Total Associations
Strategy Non-Alignment
Strategy Alignment 1 0 0 0 1 2 4
Strategy Non-Alignment
Conflict/Tension 0 0 0 0 0 0 0
Strategy Non-Alignment
Effective Communication
0 0 0 0 1 1 2
Strategy Non-Alignment
Effective Mgt Control 0 0 0 0 1 0 1
Strategy Non-Alignment
Effective Motivation 0 0 0 0 0 0 0
Strategy Non-Alignment
Ineffective Communication
1 0 0 0 2 3 6
Strategy Non-Alignment
Ineffective Mgt Control 1 0 0 0 0 0 1
Strategy Non-Alignment
Ineffective Motivation 2 0 0 0 0 2 4
Strategy Non-Alignment
Positive Outcomes 0 0 0 0 1 3 4
Effective Motivation Strategy Alignment 7 6 1 1 12 5 32
Effective Motivation Conflict/Tension 2 3 1 0 0 4 10
Effective Motivation Effective Communication
0 3 0 0 3 0 6
Effective Motivation Effective Mgt Control 8 2 1 1 14 12 38
Effective Motivation Ineffective Communication
2 3 0 0 12 9 26
Effective Motivation Ineffective Mgt Control 1 0 0 0 5 8 14
Effective Motivation Ineffective Motivation 2 2 1 0 2 5 12
Effective Motivation Strategy Non-Alignment 0 0 0 0 0 0 0
Effective Motivation Positive Outcomes 6 3 1 1 9 15 35
Ineffective Motivation
Strategy Alignment 1 1 1 2 23 10 38
Ineffective Motivation
Conflict/Tension 5 6 0 1 5 3 20
Ineffective Motivation
Effective Communication
0 1 0 0 4 2 7
Ineffective Motivation
Effective Mgt Control 0 0 0 0 5 4 9
Ineffective Motivation
Effective Motivation 2 2 0 1 6 2 13
Ineffective Motivation
Ineffective Communication
2 7 2 1 4 6 22
Ineffective Motivation
Ineffective Mgt Control 0 4 2 0 5 7 18
Ineffective Motivation
Strategy Non-Alignment 0 2 0 0 2 0 4
Ineffective Motivation
Positive Outcomes 5 1 1 0 8 10 25
continued
49
Table 4 (continued)
First Code Second Code Enclosed by
Encloses
Overlapped by
Overlaps
Follows*
Precedes*
Total Associations
Conflict/Tension Strategy Alignment 1 3 0 0 3 6 13
Conflict/Tension Effective Communication
0 0 0 1 1 0 2
Conflict/Tension Effective Mgt Control 0 2 0 0 2 1 5
Conflict/Tension Effective Motivation 4 2 0 1 3 0 10
Conflict/Tension Ineffective Communication
3 6 1 1 4 2 17
Conflict/Tension Ineffective Mgt Control 1 3 1 0 3 2 10
Conflict/Tension Ineffective Motivation 8 4 1 0 3 6 22
Conflict/Tension Strategy Non-Alignment 0 0 0 0 0 0 0
Conflict/Tension Positive Outcomes 0 0 0 0 4 2 6
Positive Outcomes Aligned with Strategy 8 9 0 1 21 11 50
Positive Outcomes Conflict/Tension 0 0 0 0 2 4 6
Positive Outcomes Effective Communication
0 2 0 0 4 5 11
Positive Outcomes Effective Mgt Control 3 3 0 1 5 5 17
Positive Outcomes Effective Motivation 2 6 1 1 13 7 30
Positive Outcomes Ineffective Communication
2 3 0 0 9 5 19
Positive Outcomes Ineffective Mgt Control 3 0 0 0 12 1 16
Positive Outcomes Ineffective Motivation 1 4 0 1 10 9 25
Positive Outcomes Strategy Non-Alignment 0 0 0 0 3 1 4
50
ENDNOTES
1
The growing body of research that has investigated empirical links between non-financial and finan cial measures of
performance in a variety of firms and industries also includes: Amir and Lev, 1996; Banker et al., 1993, 1995, 1996, 2000;
Barth, and McNichols, 1994; Behn and Riley, 1999; Foster and Gupta, 1990, 1999; Ghosh and Lusch, 2000; Hughes,
2000; Ittner and Larcker, 1997, 1998a; Perera et al., 1997]. These studies often find significant relations between non-
financial measures and measures of financial performance, though studies of the performance effects of including non-
financial measures in co mpensation plans are less consistent. Given extensive theoretical and growing empirical support,
it is not surprising that many organizations report that they are turning to forward-looking, non-financial information to
both guide decisions and evaluate cu rrent performance [Ittner and Larcker, 1998b].
2
A similar approach to combining multiple measures of performance, the tableau de bord, has been used by some French
companies for many years [Epstein and Manzoni, 1997].
3
Objective knowledge is observable and expressible in normal language – production processes and outcomes, for
example. Tacit knowledge, however, is known and understood but not easily expressed in language – an individual’s
experiences or insights, for example. This paragraph draws heavily from Tucker et al. [1996].
4
Consideration of time lags may be important to describing these cause and effect relations [e.g., Banker et al., 2000;
Norreklit, 2000].
5
For example, Banker et al. [2000] provide empirical support from extensive time-series data within a service firm for
relations between leading non-financial measures and lagging financial performance. Furthermore, they use an event-
study method to find beneficial performance effects from including non-financial measures in management performance
evaluations.
6
Proponents of economic-value added, or EVA
TM
, also claim improvements over traditional financial measures of
performance, but it, too, is a summary financial measure, albeit one that corrects for alleged financial reporting errors.
EVA
TM
does not incorporate complementary, non-financial measures of performance.
7
While the first claim for the value of multiple measures of performance would generate little controversy beyond
considerations of costs and benefits, the second claim is a bold and rigorous hypothesis. A literal and potentially testable
description of the BSC is that it describes contemporaneous, leading, or lagging relations among performance measures.
For example, improvements in learning and growth such as increased training should be reflected in predictable
improvements in internal processes, such as reduced cycle time [e.g., Luft and Shields, 1999]. Likewise, improvements in
internal processes would result predictably in improved customer value (e.g., satisfaction and market share). Finally,
improvements in customer value would lead to predictable increases in financial success (e.g., profits). Creating such a
comprehensive and coherent model is an ambitious objective that is akin to simulating the business model of the
organization itself. Accomplishing such an empirical result may not establish causality among BSC elements because (1)
some proposed measures may not be independent, (2) causes of profitability may not be generalizable beyond the
context of a specific firm [Norreklit, 2000], and (3) factors omitted from the model may be correlated with both causes
and effects.
51
8
Unless otherwise cited, this section draws from summaries in Simons, 2000: chapter 11 and Merchant, 1989: chapter 2.
9
Though this appears to be a novel application of the BSC, crossing the normal boundaries of the firm, it seems
reasonable to expect that the BSC could be used to control and communicate strategy with “business partners” as well as
normal employees. This use of the BSC may become particularly important as firms increasingly outsource more parts of
the value chain.
10
Baxter and Chua [1998] provide a thoughtful essay on the practical difficulties of conducting field research, the first of
which is gaining access to field sites. The method of access also may be a source of bias and a threat to internal and
external validity of the field research [e.g., Atkinson and Shaffir, 1998]. Though several employees that we contacted
knew or knew of the author’s relative, we had no direct contact with that relative or his/her direct reports as part of this
research effort. It is undeniable, however, that this relationship improved our access, but also may have moderated what
some individuals revealed to us. We are not able to determine the expected direct ion or magnitude of any “sponsorship”
bias, but since we heard considerable, apparently unrestrained criticism of company practices, we do not feel that any
realized bias to suppress criticism was significant.
11
Miles and Huberman [1994] observe that random sampling usually is an inefficient approach to qualitative research,
particularly when the research is theory-driven.
12
For purposes of this paper, we consider corporate citizenship as a dimension of the BSC’s customer value.
13
Weights may reflect the strength of causal links in a statistically fitted BSC, but the statistical analysis had not been
completed at the time of this study. This is an area of current research. Thus, weights reflected management’s degree of
belief about the importance and quality of each measure, as explained later.
14
The DBSC pre-dated this specific terminology and technique, but the concept of communicating the “story of
success” did exist.
15
The company has established quantitative thresholds for each color rating of each measure of the DBSC. We are not
at liberty to disclose these thresholds.
16
In figure 1 we have obscured names of the representative distributorships and have ordered them randomly. This
figure does show actual performance ratings for specific distributorships.
17
Lillis [1999] notes that “papers reporting the results of (qualitative) research studies disclose little detail regarding
attributes of study design, analytical processes and methods actually used by researchers.” Because the method of this
paper is relatively new to the accounting literature, we devote considerable space to this topic. We acknowledge that
computer-aided methods of qualitative research are controversial within their fields of origin, anthropology and
sociology. The primary source of controversy appears to be relative emphasis on positive method promoted by
computer methods versus insightful analysis allegedly sacrificed to the rigidity of the method [e.g., Coffey et al., 1996;
Lee and Fielding, 1996], but this type of controversy is familiar in most sub-fields of accounting, too. See also Trochim’s
[2000] Research Methods Knowledge Base at http://trochim.human.cornell.edu/kb/ .
52
18
The sampling scheme might be biased because of the relative under-sampling of “green” distributors. The relative
tenor of comments, however, does not appear to be related to the overall DBSC rating. There are no statistically
significant (α = 0.05) correlations among DBSC scores or “effective” or “ineffective” co mments. It does not appear that
the sampling scheme is a source of bias in subject responses.
19
Distributors claimed that they always told the company what they thought, without restraint, though many were not
sure they were heard. This is consistent with the study’s characterization of the management style as top-down.
20
One alternative approach, which is more constrained and possibly more objective, is to conduct a written or telephone
survey to elicit scaled responses to theoretically derived questions regarding specific characteristics of management
control or organizational communication [e.g., Dillman, 1978]. This approach requires a more fully developed theory of
BSC effectiveness than we believe was available and may restrict the range of data collected. The findings of this study
could be used to develop a mail or telephone survey for gathering cross-sectional data.
21
We began the first interview with nine distributor-suppled measures but quickly found it more descriptive to split
several into separate measures. The set of codes used and their definitions are in Appendix 1.
22
Figure 2 also contains examples of code associations, which are detailed in Appendix 2.
23
Conflict and tension can be beneficial when they serve as catalysts for improved communication and resolution, but
conflict that is left to simmer can worsen relations [e.g., Watson and Baumler, 1975].
24
Numbers in brackets reflect interview number and lines of text referenced, e.g., [3: 154 -158] = interview 3, lines 154 -
159.
25
It was less feasible in this study to assess plausibility, temporality, or behavioral gradient, to analogize, or to experiment
with levels of factors. See Appendix 2 for discussion of the characteristics of causal relations.
26
Numbers on the links in figure 4 are counts of the verified, paired and linked quotations, corresponding to table 3.
27
Nearly all distributors talked naturally about performance in terms of color ratings and rankings. . If nothing else, both
color ratings and rankings from this DBSC have changed the company’s language and reinforce its already competitive
culture. For example,
“I would be really reluctant to post this on the bulletin board. I don’t want customers or technicians to see red” [1: 154-
156]. “If you’re red you’re an idiot” [3: 172]. “By seeing it all, you can just call someone up and say, ‘How did you get
green in service utilization?’” [4: 188]. “We were yellow, now we’re bright red” [9: 59]. “We are competitive, so it matters
what rank you are… I want to be number one” [5: 18].
28
Most expressed uses of the DBSC (by both managers and distributors) appeared to be as diagnostic rather than
interactive controls, also consistent with the company’s top-down culture.
29
This quotation also is an example of a highly salient sub-code link (meaningful reward ? improvement) that did not meet
our frequency threshold of at least 10 associations, illustrating one of the possible costs of striving to increase the
objectivity of the analysis.
53
30
Distributors expressed strong feelings about several measures, which they felt were measured poorly, but accounted
for relatively little weight in the overall DBSC score (e.g., safety had a 2 percent weight). This probably reflects general
dissatisfaction with both the visibility of their poor performance on those measures and the top-down process of
developing the DBSC.
31
Several weeks were sufficient for this team of researchers to all but forget the details of the original coding. Because
the transcripts were relatively long and the coding was complex, there seems little chance that recoding was based on
recall of original coding.
32
We are aware of research in other fields that uses as few as seven to ten codes and reports higher coding reliability.
Though this may not be the only determinant of coding reliability, we did not reduce our coding scheme drastically just
to increase this statistic. Qualitative methods sacrifice some objectivity to gain increased relevance, but by being as
objective as possible computer-aided qualitative researchers at a minimum disclose their potential biases and create an
auditable database.
33
The choice of “one” line is discretionary and conservative. Upon further investigation, we found that codes within one
line of each other, with one exception – which was discarded, were part of the same continuous thought and were
evidence of causality; whereas many codes as much as five lines apart were coincidentally proximate – and thus not
included as evidence of causality. For ease of coding, transcripts were saved in approximately 60-character lines.
34
These attributes are similar to Einhorn and Horgarth’s [1986] spatial and temporal cues of causality.
35
Twenty-seven supercode “hits” is the mean number of supercode hits (15) plus one standard deviation (12).
