Jump to ContentJump to Main Navigation
Winning the Reputation GameCreating Stakeholder Value and Competitive Advantage$

Grahame R. Dowling

Print publication date: 2016

Print ISBN-13: 9780262034463

Published to MIT Press Scholarship Online: January 2017

DOI: 10.7551/mitpress/9780262034463.001.0001

Show Summary Details
Page of

PRINTED FROM MIT PRESS SCHOLARSHIP ONLINE (www.mitpress.universitypressscholarship.com). (c) Copyright The MIT Press, 2017. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a monograph in MITSO for personal use (for details see http://www.mitpress.universitypressscholarship.com/page/privacy-policy). Subscriber: null; date: 26 March 2017

Measuring Corporate Reputations: Keeping Score

Measuring Corporate Reputations: Keeping Score

Chapter:
(p.139) 8 Measuring Corporate Reputations: Keeping Score
Source:
Winning the Reputation Game
Author(s):

Grahame R. Dowling

Publisher:
The MIT Press
DOI:10.7551/mitpress/9780262034463.003.0008

Abstract and Keywords

This chapter explores the old saying that “what is measured improves”. It presents different ways that corporate reputations are measured and some new ways that may be used in the future.

Keywords:   Best-worst measures, Media analysis, Off-the-shelf measures, Customised measures

What is measured improves.

Peter Drucker

and also

  • A firm becomes what it measures.

John Hauser and Gerald Katz

To prime the material in this chapter, I will illustrate how to measure whether an organization has a better corporate reputation than its rivals. The task I assume is to find a “best–worst” measure of the corporate reputations of a number of long-haul airlines that fly out of Australia. As I will elaborate later in the chapter, this measure is in fact a more scientifically valid way to measure the relative corporate reputations of a group of companies than any opinion poll approach such as that adopted by Fortune magazine to produce their annual ranking of the World’s Most Admired Companies.1

For this task I have chosen seven competitor airlines. My evaluation of these airlines was done after Malaysian Airlines had suffered two major catastrophes, namely, Flight MH370 that went missing from Kula Lumpur to Beijing and Flight MH17 that was shot down over east Ukraine. If you follow the instructions in the question below and put in your choices, you would see how your evaluation differs from mine. The airlines are:

  • Qantas

  • Emirates

  • British Airways

  • (p.140) Singapore Airlines

  • United Airlines

  • Thai Airlines

  • Malaysian Airlines

In the question below you need all seven airlines because this measure is based on an experimental design for seven airlines. (The same can be done for any number of companies, but it requires a different experimental design for each number.)

Question

Please evaluate the overall corporate reputations of the seven airlines listed above in accord with the admiration and respect you hold them in at the present time. This evaluation should be based on what you know about the airlines. You may have gained this knowledge by flying with them, seeing their advertising, listening to what the media is saying about them, or talking to other people.

Your Task

For each of the sets of airlines below please select the one with the best and the one with the worst overall reputation from each set. (p.141)

Set 1

Best reputation

Worst reputation

Qantas

Emirates

British Airways

Set 2

Best reputation

Worst reputation

Qantas

Singapore Airlines

United Airlines

Set 3

Best reputation

Worst reputation

Qantas

Thai Airlines

Malaysian Airlines

Set 4

Best reputation

Worst reputation

Emirates

Singapore Airlines

Thai Airlines

Set 5

Best reputation

Worst reputation

Emirates

United Airlines

Malaysian Airlines

Set 6

Best reputation

Worst reputation

British Airways

Singapore Airlines

Malaysian Airlines

Set 7

Best reputation

Worst reputation

British Airways

United Airlines

Thai Airlines

Scoring

Step 1. Add up the number of bests and worsts for each airline and then compute a net score (best to worst).

Qantas

Bests: 1

Worsts: 0

Net score: 1

Emirates

Bests: 2

Worsts: 0

Net score: 2

British

Bests: 0

Worsts: 1

Net score: −1

Singapore

Bests: 3

Worsts: 0

Net score: 3

United

Bests: 0

Worsts: 2

Net score: −2

Thai

Bests: 1

Worsts: 1

Net score: 0

Malaysian

Bests: 0

Worsts: 3

Net score: − 3

Step 2. Rank the companies from highest to lowest scores—remember that the bigger the negative score, the worse the reputation.

  1. 1. Singapore (+3)

  2. 2. Emirates (+2)

  3. 3. Qantas (+1)

  4. 4. Thai (0)

  5. 5. British (−1)

  6. 6. United (−2)

  7. 7. Malaysian (−3)

And that’s it. You have a scientifically valid rank ordering for you and me of where each airline’s reputation sits against these six competitors.

(p.142) Why this process creates a valid and reliable measure of corporate reputation is that it is based on a formal theory of choice where people prefer the best over the worst option. And if corporate reputation is to drive competitive advantage, the measure needs to reflect this. In this case choosing an airline is a better indicator of competitive advantage than rating the airline on a scale ranging from say “highly preferred” to “not at all preferred.” The measure is also based on an experimentally designed set of alternatives to consider.2 These are called balanced incomplete block designs. In the design above every company is compared once with every other company, which means that it appears in three choice sets.

The other advantage of this measure is that it is really easy for most people to do because they have an intuitive notion of what a good or bad reputation is and because most times, when they fly, they choose one of the airlines from the set of those going to that destination. The disadvantage of this type of measure is that it is not diagnostic because it does not provide any information about why the respondent thinks that each airline has a good or bad reputation. However, a couple of follow-up questions can gather this information.

The diagram of corporate reputation formation in chapter 3 suggests that a company’s reputation is based on the salient beliefs people hold of the organization. To help discover these, most big companies routinely analyze the media in which the company is reported. The corporate affairs department and the CEO receive a regular flow of media analysis reports. Many companies also mine their qualitative customer research to discover this group’s beliefs about the company. Customer satisfaction surveys are often used as a surrogate for reputation.3 For employees, the annual climate, culture, and satisfaction surveys can be helpful. For other stakeholders, most companies rely on ad hoc surveys and episodes of hostile feedback to gain insight into the strength of the reputation. Sometimes companies will commission specific reputation research from an organization like the Reputation Institute. This organization will do “deep dive” analysis of the company’s scores on the twenty attributes or belief statements used to profile its reputation. And if asked, it will use the same measure across competitors. If this mixture of data across the company sounds a bit ad hoc, it is because it is!

On the academic side of the measurement ledger, the situation is not much better. Naomi Gardberg and I spent a year reviewing a variety of (p.143) scholarly measures of corporate reputation only to conclude that while they are improving, a lot of work still needs to be done in order to produce a valid and reliable measure of corporate reputation.4 Because most of these measures are designed around a specific research question, they are not flexible enough to provide a good measure of the corporate reputations of rival companies across multiple stakeholder groups.

While this is a fairly bleak picture of the current state of affairs, it need not be. In this chapter I’ll describe what a good measurement landscape should look like. And I’ll identify who can help gather the data and develop sound measures of corporate reputation. Because nearly all companies do media analysis, I’ll start here. Then I’ll describe how to develop a set of customized measures of corporate reputation that are useful for diagnostic purposes and for determining if the company has won or lost the reputation game.

What the Media Is Telling You about Your Corporate Reputation

This section is based on my association with Warren Weeks, the founder of Cubit Media Research.5 Our collaboration grew out of Warren’s frustration with many of the media summaries given to CEOs. Most consisted of counts of positive and negative mentions of the company and its products tagged against different media, and sometimes incidents involving the company. This is a crude form of sentiment analysis. Through the lens of corporate reputation it is only marginally helpful.

To put media coverage in context, it is useful to note that the business media where companies are talked about have developed their own vernacular and way of framing issues. Of importance here is their focus on share price and events that “move markets.” Moreover many outlets are ideological and will push the idea of free-market capitalism or sometimes social responsibility.6 Within this context, there are four major roles that the traditional media serve when they report on firms:

  • To report on significant events that occur in the business and economic environment. As we saw during the global financial crisis, the way that these events are reported and dramatized can affect public perception about the role and contribution of business.

  • (p.144) To report about the strategies, financial performance, employee layoffs, new product introductions, and the like, of specific companies. The positive and negative mix of these can taint a company’s reputation. And it can help position the company as best at something or best for somebody.

  • To act as an independent investigator. Here journalists are creating news and acting as a watch-dog for society. Many of these stories have a negative tone of voice and thus may adversely impact the corporate reputations of the companies targeted.

  • To provide a platform to publicize the views of a group of external stakeholders. Here they act as a messenger to company leaders. They also provide a degree of extra legitimacy to the message and the groups involved.

The rise of the Internet media has added a few new twists to reputation monitoring. Now it is not uncommon to find hate sites, spoof and parody sites, NGO sites, and anti-corporate sites. For example, Greenpeace helped orchestrate a campaign against Shell’s Arctic oil drilling campaign. The campaign had a fake video, a fake press release, and a fake website. A few months after its creation, it had attracted 785,000 views and 1,600 comments.7 What actual effect this site had on Shell’s behavior or its reputation is hard to determine. While the number of views and comments look impressive, it is difficult to know if Shell took them seriously and/or made any changes to its fundamental operations that it otherwise would not have carried out. The reason for this assessment is because Tom Bower describes Shell as a very hard-nosed and arrogant company. Thus its reputation suggests that one episode of public screaming would be unlikely to significantly affect its plans.8 Sometimes funny and not-so-funny videos featuring a company or its products will be uploaded to YouTube. These websites can also attract a large number of followers, and occasionally they spark a viral social network campaign. Some media-monitoring firms specialize in tracking these sites. There is a wonderful irony here. The successful Internet sites are often monitored by journalists who then go on to introduce the issues they find to the more mainstream media. This provides the original story with more credibility.

One of the reputation problems with social media is that this medium generates a lot of noise. That is, as the volume of social comments increases, so does the amount of chatter between people arguing with (p.145) each other rather than generating commentary that is focused on the company and its reputation. Thus it becomes more difficult to distinguish genuine concern, or signal, from argumentative chatter, or noise. As the ratio of signal to noise falls, it becomes harder to get a clear picture of the company. Paradoxically, this makes it easier for poor companies to hide in the morass of social chatter.

Over time media coverage defines what is important for many people to believe about companies and what aspects of corporate character and performance are being used to evaluate them. For example, some years ago one of the big European technology companies wanted to project an image as a technology leader, but its press coverage consistently portrayed it as selling stylish products. One of the reasons for this discrepancy was that the business journalists thought that while the company had good technology, which they framed as “stylish,” they also thought that it was slow to get its products to market. Hence it was not a “leader.” Thus the company developed a media reputation for product design rather than innovation. Not bad, but not what they wanted.

To understand the nature of a company’s media reputation requires two things. The first is the type of story the journalist is pursuing. For example, is this a story about a company that:

  • is on top, or

  • perched for a fall, or

  • struggling to survive, or

  • in a turnaround situation, or

  • in the middle of a crisis?

And is the journalist a self-proclaimed reporter or do they have an agenda, such as being a “professional destroyer of reputations” as one journalist described himself on a YouTube segment.

The second issue is to analyze the words and phrases used by the journalists and bloggers to describe the company and its competitors. Because the words used by people are the ones that have real meaning to them, the number of times a word is used will be a good guide to its importance. Some of these words will fit neatly into the natural language of reputation and how the company would like to be evaluated. Other words will be new and insightful. And sometimes the writer will reveal the values that he or she is using to judge the morality of the company’s behavior. For (p.146) example, during the extensive media coverage of the Qantas industrial dispute referred to earlier, the CEO was frequently referred to by his name Alan Joyce, not his role as CEO. Thus the story became personal. It was often framed as Alan Joyce of Irish heritage who was leader of an Australian icon company fighting with his Australian workforce and the person directly responsible for the hardship to Australians Mary and Peter Traveling Citizens stranded in a foreign country. You can imagine the emotion charged watercooler conversations about how to fix the Qantas dispute. And from then on for many people the reputation of Qantas is affected by the reputation of Alan Joyce the Irish-born CEO.

As the previous examples illustrate, the analysis of words and phrases should be done to reveal the themes, contradictions, expectations, and values in what is said. This is important not only to highlight any misalignment between a company’s desired and media reputations but also because people seldom unconditionally like or dislike a company. Discussion about the company can now move along from “we have a good or bad reputation,” to “we are known to be good at this and not so good for that.” The contradictions between good and bad aspects of a company are insightful because they demand an explanation.

To illustrate these ideas, consider an article about Apple published in Business Week titled “A bruise or two on Apple’s reputation.”9 The article contains several elements:

  • Business Week story line—“Is the company’s stellar service keeping up with its hyper growth? Some customers don’t think so.”

  • Journalists’ themes—as Apple’s new products (iPod, iPhone) become more successful, they are being purchased by customers who are less devoted to Apple and who are often less “tech-savvy.” “The vitriol of complaints on some Apple -related blogs and websites is approaching that usually reserved for cable TV.” Positive endorsement from a long-time customer Nigel Ashton. Negative endorsements from two new customers Catherine Temple and Michael Levin accompanied by their forlorn photos.

  • Apple’s position—Timothy Cook claims that an array of internal metrics show service has never been better.

  • Contradictions—Cook versus the two new customers and the blogs. Cook versus an academic expert who endorses the claim that as the customer base becomes more diverse it becomes harder to satisfy.

  • (p.147) Comparisons—“Even small cracks in a pristine reputation can be a sign of larger problems. Just ask Dell.” Table of customer satisfaction scores from an independent research firm shows Apple (79 percent down from 83 percent), HP (76 percent up from 75 percent), Gateway (75 percent up from 73 percent), and Dell (74 percent down from 78 percent).

The rich content of this three-page article is lost if it is classified as either a mostly positive or mostly negative or a mixed piece of journalism. However, if it is examined through a multi-focal lens, then we see strategic issues (market expansion), industry issues (all competitors have 75 percentage scores on satisfaction), and product and service issues (new multi-function products make it harder for some customers and company service people to get these products to work). Some of these issues support the Apple corporate story of great product design and some challenge aspects of it. Also we see conflict—Cook contradicting the journalists’ evidence in the article. This “you say” versus “we say” piece of journalism is likely to be noticed by employees. Hence it will require a clear response to them from senior management.

When journalists and bloggers write about a company, they expose their mental model of what is important for success in an industry. For example, John Gapper states that “when a company is doing noticeably better than competitors in its industry, there are three possible explanations: skill, luck, or edge.”10 Media coverage often reflects the journalist’s view about which of these is operative for the company in question. These mental models can be better understood by interviewing the people who follow and talk about a company. For example, reading the financial press suggests that a company will be respected if it is a good exponent of the strategy of exploration or exploitation. Exploration is based on innovation, whereas exploitation is based on becoming more efficient. However, to fully understand how a journalist calibrates a company’s implementation of either strategy requires understanding which corporate behaviors they focus on to signal each strategy. To launch innovative new products may signal a good explorer company. Cost cutting, restructuring, and the adoption of modern management techniques may signal a good exploitative company. Understanding these mental models then helps the company to tailor its communication to these opinion leaders.

(p.148) Based on the analysis of tens of thousands of media pieces, Cubit Media Research has discovered that most can be classified into one of eight broad macro themes. These are shown in figures 8.1 and 8.2. The business themes tend to be described with reference to “hard numbers,” while the social themes are often written about in terms of corporate behaviors that impact on specific types of stakeholders. The business themes tend to occur more in weekday media, while the social themes in weekend media. Each macro theme is comprised of a number of micro themes. For example, financial performance may be discussed in terms of earnings, profit, strategy, growth, and the CEO. Social themes often contain human interest stories and product themes such as new products. In the “old media” journalists tend to specialize in industries.

These two figures summarize the numbers behind the media coverage. They act as starting points for a discussion of a company’s media reputation. Figure 8.1 shows company salience, namely which companies are getting coverage. The media play an important role here. Who they select for attention and what they say puts these companies’ reputations in play. This figure also shows whether all the companies are being talked about in the same way. For example, our company’s products and services are not nearly as noteworthy as those of our major competitor.

Figure 8.2 shows the risk profile of the media coverage—what topics are attracting positive, mixed, or negative stories. Here we see that regulatory issues seem to be attracting too much negative attention. Having a public argument with a regulator or being put under investigation are common causes of such media activity. They may also trigger a values-based discussion of the company about its ethics or moral responsibility. Another worrying feature of this graph is the negative commentary about stakeholders. Here stories about disgruntled stakeholders can infect other people. The field of word-of-mouth and viral marketing suggests that negative commentary can be more damaging than the boost provided by a similar amount of positive coverage.

The numbers in the two figures are invitations to unpack each macro theme into its micro themes. For example, do the product and stakeholder themes focus on the superior or inferior value offered to these people? Also, when journalists focus on strategy and governance, one micro theme that has proved troublesome for many companies is the profile of the (p.149)

Measuring Corporate Reputations: Keeping Score

Figure 8.1 Media salience

Measuring Corporate Reputations: Keeping Score

Figure 8.2 Our company’s profile

(p.150) CEO. In some countries, charismatic leadership is a positive theme, while in others, it is tainted with celebrity. In either case, when a high-profile CEO is called to account in the media, their reputation and that of their company is in play. The effects of this can be particularly damaging inside the company. After a series of such encounters with the media one high-profile CEO of a US company was chastised inside her company with the chant of “ditch the bitch.” She was eventually ditched by the chair of the board who soon after suffered the same fate!

The next part of media analysis is to relate each macro theme to the language of corporate reputations. Are values-laden words used? Are stakeholder value words used? Are reputation words such as admiration, respect, trust, reliability, credibility, confidence, honesty, status, fairness, and legitimacy also used when discussing corporate issues? And why is this language used? For example, a company will lose credibility when it breaks a promise. It will be judged as unreliable if it fails to meet expectations. It will lose its moral authority if its behavior doesn’t fit with the values of its stakeholders. In essence a company needs to know if the media is talking about its reputation. Very few conventional media studies are structured to highlight the links between what journalists and other opinion leaders say about a company and its reputations.

The final part of media analysis is to think about how to act on the information uncovered. For example, positive message themes need to be protected, and other “missing” positive themes need to be advanced. Often a well-known theme can be used to provide credibility for a missing theme. As noted earlier, GE has two such positive message themes. One is its well-touted record of profitability, and the other is the company’s less talked about environmental (Ecomagination) initiative. By setting specific financial targets for its Ecomagination products and services GE could link its profit story to its environment story. For a negative message theme, two options are available. One is to fix the underlying issue or to communicate how it is being addressed. The other is to seek to correct the misconception by providing new information or briefing selected journalists. For a mixed message theme, the task is to understand whether this results from a contradiction of what the company is saying or inconsistent messages. Then the task is to promote your side of the story.

(p.151) Developing Customized Measures of Corporate Reputation

I now shift attention to describing three quantitative measures of corporate reputation. As noted in the quotes at the beginning of the chapter—“If you can’t measure it, you can’t manage it.” Hence we need some measures. BUT!! It is important to understand the nature of any measure chosen to calibrate a corporate reputation. In the following discussion I’ll avoid a technical discussion of this because it is a bit complicated, but I’ll highlight some of the commonsense aspects of reputation measures.

The three measures described below are interrelated. The role of the first measure is to calibrate the rank order of a company’s reputation against its rivals. This will indicate the potential for the company’s reputation to lead to a competitive advantage. The role of measure 2 is to discover the basis by which people evaluate the company’s reputation. The role of measure 3 is to calibrate how much a good reputation is worth in a specific situation. The central idea of this measure is to determine the relative importance of a good or not so good corporate reputation relative to the other attributes people use to make their choices about which companies to engage with.

Measures That Matter 1: Is a Company Better or Worse Than its Competitors?

Recall the best–worst measure of reputation in the introduction of this chapter. This is a good example of a quick and psychometrically valid measure of how one organization’s reputation compares with its competitors. As noted earlier, measures of different numbers of organizations will require different experimental designs. Any market research firm that does choice modeling can design these.

A critical part of this style of measure is to explain what you mean by the concept of corporate reputation to the people rating the organizations. The failure to do this is one of the biggest problems with the most prominent measure of corporate reputation in the scholarly literature. It is Fortune magazine’s measure of the World’s Most Admired Companies. In this measure each respondent is asked to rate a number of companies on nine attributes, namely ability to attract and retain talented employees, quality of management, social responsibility to the community and the environment, innovativeness, quality of products and services, wise use of (p.152) corporate assets, financial soundness, long-term investment value, effectiveness of doing business globally. However, nowhere in the instructions to respondents is the notion of reputation explained to them.11 Hence each respondent is free to interpret each of the nine attribute rating scales as they wish. Thus when you add up these nine scores, each rated on an 11-point scale to get an overall evaluation, what you get is a very messy, or as some have labeled this type of measure as “junk.”12

One of the critical advantages of the best–worst type of measure at the beginning of this chapter is that it forces the researcher to define what reputation means before asking people to rate the reputations of a set of competitive companies. And because the choice options for each set of companies are best reputation and worst reputation, the question format captures the semantic understanding of reputation that most people have. Thus reputation is a much more commonsense construct to measure than something like “the wise use of corporate assets.”13

Measures That Matter 2: What Makes a Company Good and Bad?

A number of times I have mentioned that people seldom unconditionally consider a company to be all good or all bad. And as explained earlier, this is more likely to be the case for people who are better informed about the company. The insight that can be gained from knowing what is good and bad about a company, however, is lost in most quantitative measures of corporate reputation, both those of academics and practitioners. The reason is that they create single-number summary measures that are easier for the media to report and much easier for the statisticians to correlate with other variables. For example, in the table of 2012 RepTrakTM winners in chapter 10 it is easy to see that Rolls Royce has a better reputation than Specsavers. In the report where these numbers appear each company’s score is correlated with the “percentage of respondents who would recommend that company.” The correlation is 0.79, which is very high. But what does this mean?

This correlation can mean either that

  • as a company’s reputation increases, so does the percentage of people who will recommend it increase, or that

  • as the percentage of people who recommend a company increases, so will its reputation increase.

(p.153) Each interpretation is equally valid because a correlation tells you nothing about the direction of causation. And in this case it can be argued that a good reputation causes positive recommendations about the company AND that positive recommendations by people enhance the reputation of the company in question.

These Rolls Royce and Specsavers reputation numbers are the aggregate scores of 20 questions, each measured on a 5-point scale. However, what these total scores hide is that fact that two companies with the same score can have very different corporate reputations. This will occur in the middle of the range of scores, precisely where most companies will be located. Figure 8.3 illustrates this point. To keep it simple, I present three companies whose reputation is measured on four 5-point scales. Each has a very different reputation profile yet each has the same overall reputation score. Thus the way that the measure is communicated in the polls camouflages the structural differences between the reputations of these companies. When this happens, the overall reputation score really is junk.

Throughout the book I have described three competing models of reputation formation, namely the stakeholder value proposition (SVP) model, the character and competence model, and the corporate social responsibility (CSR) model. Now which of these is adopted will

Measuring Corporate Reputations: Keeping Score

Figure 8.3 Similar scores but different corporate reputations

(p.154) determine what is included in a measure of corporate reputation. For example:

  • The SVP model assumes that being evaluated as best at something and/or best for somebody is the most important driver of a strong and distinctive corporate reputation. Thus SVP-based measures of corporate reputation would be made up of questions about the attributes of the value proposition offered to each particular group of stakeholders, such as those in the “onion diagrams” described in chapter 6. The typical way of doing this is to get respondents to rate the importance of each attribute’s contribution to having a reputation for being known as best at something and/or best for somebody. Five or seven-point Likert-type rating scales could be used. The problem with this method, however, is that if prior research has been used to help identify the important attributes, then one should expect that most respondents will rate most of the attributes as important. And they do tend to do this. Hence this approach is not very helpful as a way to find out what really matters. A better approach is to ask respondents to rank-order the attributes from most to least important.

The SVP approach to corporate reputation measurement would result in a different measure for each stakeholder group. Because each is composed of a different set of attributes, these reputation measures would not be comparable.14 By this I mean that from a measurement theory perspective, it would not be meaningful to state that the company has say a better reputation among its employees than its customers. These SVP measures provide a fine-grained understanding of the corporate reputations of various stakeholders groups, but this comes with the restriction that they are noncomparable and they can’t be added together to get a measure of the company’s overall corporate reputation. For many companies this will not be too burdensome because they are interested in their corporate reputation in different markets, such as the market for employees or the market for customers.

If a company does want to compare its SVP-based corporate reputations across groups, it could nevertheless do this by asking each group to also rate the overall reputation of the company as follows:

Please evaluate the overall reputation of Company X. By corporate reputation, I mean the admiration and respect you hold this company in (p.155) at the present time. You can rate Company X as predominantly good or bad, or if there are significant elements that are both good and bad, then please rate it as mixed. If you don’t know much about this company then please tick the “don’t know” option. If for whatever reason you “don’t care” about this company or its reputation, then please select this option.

Good

Bad

Don’t know

Don’t care

+2

+1

−1

−2

There are a couple of important elements of this measure. One is that there is no midpoint option on the answer scale such as a 0. This is to get the respondents to make a judgment. If a zero is included on the scale, research has shown that many people will select it.15 A second element is that the “good” and ‘bad” answer options are coded with positive and negative numbers. This is to signal that good is “good” and bad is “bad.” Had the scale been numbered say from 5 to 1, then this would signal that “bad” is really just “not so good.” A third element is that things are explained to the respondents. Corporate reputation is defined in the question and what is meant by “mixed”, “don’t know,” and “don’t care” is also explained. This is done to minimize the chance that each respondent will interpret the question differently.16 Finally a “don’t care” answer option is included because it is important to know if the respondent simply doesn’t care about the company. When the general public is asked to evaluate a disparate list of companies, this is a very important inclusion. It tells the researchers about the relevance of each company to the sample of people polled for their opinion. Before a company bases its reputation management on the responses of a group of people, it would be nice to know if they are both knowledgeable about it and care about it.

  • The character and competence model assumes that a set of these attributes define the identity and reputation of the company. Because these speak to all groups of stakeholders, only one set of these need to be assessed, and hence only one measure is needed.

  • The CSR model assumes that being evaluated as a good corporate citizen is the most important driver of a good corporate reputation. (p.156) Here the relative contribution of various attributes of CSR to the reputation of the company would be measured. Again, because the same attributes of CSR would be meaningful to all the groups of stakeholders, only one measure of corporate reputation would be needed.

Some researchers use what I call a “mixed measure” of corporate reputation. Because they are not quite sure exactly what contributes to the overall reputation of a company, they combine SVP attributes, organizational identity attributes, and CSR attributes in the same measure.17 This is the style of measure adopted by Fortune World’s Most Admired Companies measure and the Reputation Institute’s RepTrakTM measure. These multi-attribute measures are typically composed of between nine and twenty rating scales. Here the company is assumed to have an overall reputation based on the evaluation of a very broad array of its characteristics. These measures also assume that the respondent knows a lot about the companies being rated.18 And, because the scores on the rating scales are added together or averaged to get an overall score, it is further assumed that the good aspects of the company can compensate for the bad aspects. This may not be a sensible assumption if people form their evaluations primarily by judging a company on only one or two criteria. For example, if a person decides that any company that is involved in tax avoidance is bad, then it does not matter what else it does.

These different measures highlight a fundamental difference in the way that scholars, and possibly managers, think about what a corporate reputation is. The SVP measure focuses on being good for somebody. The character and competence model focuses on being good at something. The CSR-based measure focuses on being a good corporate citizen. In contrast, the mixed measure is unfocused. It assumes that a company has a “meta” or “halo” reputation that makes it on average an admired and respected company. As noted in chapter 1, the academic community has yet to reach a consensus about which model and measure of reputation is best. However, what can be said about the four approaches is that the first three are much clearer about how a strong and distinctive corporate reputation is formed and should work than the third approach. And because of this it should be easier to measure these (p.157) effects and easier for managers to decide if they want to invest in creating a better corporate reputation.

Measures That Matter 3: Is Your Corporate Reputation Worth Anything?

The answer to this question is revealed if people place a monetary value on the company’s reputation when they make decisions about whether or not to engage with it. Surprisingly, few studies of corporate reputation actually measure this effect. They measure many things that are related to the effect, but not whether people are actually prepared to pay for the privilege of being associated with a company they admire and respect. For example, a typical empirical study will search for statistically significant correlations among (1) various corporate characteristics and the company’s overall reputation; (2) corporate reputation and trust, admiration, respect, and the like; and (3) corporate reputation and stakeholder behaviors such as customer loyalty. These relationships are important, but they don’t tell the company if its reputation is more financially valuable than those of its competitors.

What would be ideal to know is how much money a person would be willing to pay in order to deal with a company when all other aspects of their relationship are held constant. If this amount of money is larger than the amount they are willing to pay to deal with the company’s rivals, then it is winning the reputation game. It turns out that it is possible to estimate these amounts. Let me explain.

A couple of years ago my colleagues and I did a study that focused on how companies can win the war for talented employees.19 In this study we estimated that for MBA students about to enter the workforce, the value of a potential employer’s good corporate reputation was $12,388. We did this by asking each student to evaluate 16 pairs of hypothetical job contracts that were made up of features such as contract length, reputation of the company, pension contributions, promotion opportunities, salary, and travel demands. Across the contracts, each feature was varied across a number of levels. For example, the MBA salaries ranged from $90,000 to $150,000. The corporate reputation levels ranged from being in the top 25 percent globally, as named by the Reputation Institute, to not listed by the Reputation Institute but (p.158) known to receive negative media coverage from time to time. For each pair of contracts the student chose the one they most preferred. From the 16 choices each person made we statistically estimated the relative value (aka utility) of each feature. When one of the features is stated in a dollar amount, as done here with the starting salary, then with some further statistical estimation, it is possible to put a dollar value on all the other features that is anchored to the average dollar amount (of the starting salary in this case). This is where the $12,388 dollar value of corporate reputation comes from. If the salary range was different, then the value of the reputation would also be different.

The research technique we used is known as a “discrete choice experiment.” In these experiments people are not asked to rate things, they are asked to make a choice—in this case it was for a job contract (or employee value proposition) where one of the features is reputation. This is exactly what people do when seeking employment. And it is exactly what other key stakeholders such as consumers, investors, business partners, and local communities do when they engage with companies. They make choices that involve trade-offs across the features of the overall offer the company makes.

The greatest strength of this way of measuring the value of a corporate reputation is also its greatest weakness. It is far more complicated to design, administer, and produce the statistical estimates of utility than the traditional opinion poll style of research. But, after we did our employment study with the MBAs, we repeated it with samples of white-collar, medical, legal, public service, and manual employees. And they all survived the experience!

To summarize, measuring corporate reputations is not straightforward. Off-the-shelf measures like Rep TrakTM have severe limitations while customized measures mean that they have to be developed with the help of a good market research firm. My argument is that if trading on the company’s reputation is going be a key part of its strategy, then the company can’t afford not to develop a suite of customized measures. To demonstrate that this really is a simple decision, I provide a thought experiment in appendix A to show how to estimate the worth of doing customized research.

(p.159) Context Matters

Probably the most overlooked issue regarding the measurement of corporate reputations is that of context. By this I mean:

  • type of stakeholder or spectator,

  • situation in which the person is considering engaging with the company, and

  • the way questions are asked.

Each of these contexts can have a profound effect on the measured scores of the people chosen to evaluate a company. Hence each must be considered or controlled for in any valid measure of corporate reputation. They seldom are. Let me elaborate.

The beauty contests opinion poll measures of corporate reputation are essentially context free. The instructions to respondents seldom ask respondents to specify their knowledge of and relationship with the companies under evaluation. Also they seldom ask them to consider the reputation of each company as they would if they were going to say, buy a product from them for a specific use occasion, or change jobs within the next three months to advance their career, or invest $10,000 in their shares to add to their superannuation fund. Each such context can trigger a different set of salient beliefs about the company; it can change the relative importance of these beliefs, and it can evoke different values against which to judge the company. When a large number of people evaluate a company with reference to many, many different contexts, the scientific properties of the resulting measure are questionable.

Rupert Younger, currently the director of the Oxford Centre of Corporate Reputations, illustrates the importance of considering context by noting that many MBA students consider that the investment bank Goldman Sachs has a good reputation for the first job after graduation—because of the intense and varied on-the-job training if offers; but the company has lousy reputation for a long-term career—because of the intense and varied work required. So this company has both a good and a bad reputation as an employer among the same group of people—which one depends on the situation.

(p.160) One final example of how context affects reputation was recently published in the journal Science.20 It deals with the ratings many of us use on social network sites like Trip Adviser. Here we rely on digitized, aggregated opinions about companies to make decisions. Based on a series of experiments, it was found that if the initial evaluation of a business issue people saw was positive, then this created a significant positive social influence bias (+ 25 percent) that persisted over a five-month time frame. If it was negative, then no such bias was observed. Hence companies would do well to manipulate their online reputations by making sure that the first evaluation people see is positive.

Notes:

(1.) See, for example, G. R. Dowling and N. A. Gardberg, Measuring corporate reputation, in T. Singer (ed.), Sustainability Matters 2014 (New York: The Conference Board, 2014), 64–69.

(2.) See, for example, J. Louviere, I. Lings, T. Islam, S. Gudergan, and T. Flynn, An introduction to the application of (case 1) best–worst scaling in marketing research, International Journal of Research in Marketing 30, 2 (2013), 292–303.

(3.) There is often a trap with using customer satisfaction as a surrogate for reputation. If all competitors offer a similar level of service or similar quality products, then customers come to expect that this is the only thing on offer. Surveys will reveal that most of the customers will be satisfied with what they receive because there is no point in changing supplier. I have seen different surveys of the same people where, in one, they say they are satisfied with the products and service but, in the other, they say they do not hold the company in high regard.

(4.) G. R. Dowling and N. Gardberg, Keeping score: The challenges of measuring corporate reputation, in M. Barnet and T. Pollock (eds.), Oxford Handbook of Corporate Reputations (Oxford: Oxford University Press, 2012), 34–68. For another review, see S. L. Wartick, Measuring corporate reputation: Definition and data, Business and Society 41, 4 (2002), 371–92.

(5.) G. R. Dowling and W. Weeks, What the media is really telling you about your brand, MIT Sloan Management Review 49, 3 (2008), 28–34.

(6.) A. Schiffrin, The US press and the financial crisis, in A. Schiffrin (ed.), Bad News: How America’s Business Press Missed the Story of the Century (New York: New Press, 2011), 1–21.

(7.) E. Steel, How Shell was hijacked by a new style of cyber protest, Financial Times (June 22, 2012), 10.

(8.) T. Bower, Oil: Money, Politics and Power in the 21st Century (New York: Grand Central Publishing, 2009).

(9.) L. Lee and P. Burrows, A bruise or two on Apple’s reputation, Business Week (October 22, 2007), 81–83.

(10.) J. Gapper, Goldman’s glory may be short-lived, Financial Times (December 6, 2007),

(11.) Hay Group (2013). See www.haygroup.com/ww/best_companies/index.aspx?id=155). Accessed November 1, 2014.

(12.) J. Jacoby. Consumer research: How valid and useful are all our consumer behavior research findings? A state of the art review, Journal of Marketing 42, 2 (1978), 87–95.

(13.) The one thing that potentially saves the integrity of the WMAC measure is that the people who are chosen to rate the companies are all managers and (p.242) company analysts. They presumably have a similar notion of what each of these attributes means, and they believe that these are the true underlying determinants of admiration or reputation.

(14.) The issue here is equivalent to determining what is called the gamma change between the measures of a variable. See G. R. Dowling, The alpha, beta, gamma change approach to measuring change and its use for interpreting the effectiveness of service quality, Australian Journal of Management 26, 1 (2001), 55–67.

(15.) S. Dolnicar and B. Grün, How constrained a response? A comparison of binary, ordinal and metric answer formats, Journal of Retailing and Consumer Services 14 (2007), 108–22; S. Dolnicar, Asking good survey questions, Journal of Travel Research 52, 5 (2013), 551–74.

(16.) When this happens, the random error part of the measure increases, and thus decreases the accuracy of the measure.

(17.) If this type of measure is adopted then a good way to introduce the SVP attribute is to state it as the center part of the SVP onion. For example, for Riedel’s customers this would be how their wine glasses help produce a better drinking experience.

(18.) As chapter 3 suggests, these people may also have a quite complex structured reputation of the company.

(19.) P. Auger, T. Devinney, G. Dowling, C. Eckert, and N. Lin, How much does a company’s reputation matter in recruiting? MIT Sloan Management Review 54, 3 (2013), 79–88.

(20.) L. Muchnik, S. Aral, and S. J. Taylor, Social influence bias: A randomized experiment, Science 341 (August 9, 2013), 647–51.