Identify, quantify, and manage the risks to your company’s reputation

long before a problem or crisis strikes.

REPUTATION
and Its Risks
by Robert G. Eccles, Scott C. Newquist, and Roland Schatz

EEXECUTIVES KNOW THE IMPORTANCE of their companies’ repu-

tations. Firms with strong positive reputations attract better
people. They are perceived as providing more value, which
often allows them to charge a premium. Their customers are
more loyal and buy broader ranges of products and services.
Because the market believes that such companies will deliver
sustained earnings and future growth, they have higher
price-earnings multiples and market values and lower costs
of capital. Moreover, in an economy where 70% to 80% of
market value comes from hard-to-assess intangible assets
such as brand equity, intellectual capital, and goodwill, orga-
nizations are especially vulnerable to anything that damages
their reputations.
Blair Kelly

104 Harvard Business Review | February 2007 | [Link]

YEL MAG CYAN BLACK
Reputation and Its Risks

Most companies, however, do an inadequate job of man- reputational risks. That’s mainly because of the difficulty of
aging their reputations in general and the risks to their rep- factoring them into capital-adequacy requirements, most
utations in particular. They tend to focus their energies on banking-risk professionals would say.
handling the threats to their reputations that have already Given this lack of common standards, even sophisticated
surfaced. This is not risk management; it is crisis manage- companies have only a fuzzy idea of how to manage reputa-
ment – a reactive approach whose purpose is to limit the tional risk. A large U.S. pharmaceutical company reflects the
damage. This article provides a framework for proactively current state of practice among well-run organizations. It has
managing reputational risks. It explains the factors that af- an ERM system for managing operational and financial risks,
fect the level of such risks and then explores how a company as well as hazards from external events such as natural disas-
can sufficiently quantify and control them. Such a process ters, that is loosely based on the COSO framework. The firm’s
will help managers do a better job of assessing existing and vice president of risk management oversees the system. How-
potential threats to their companies’ reputations and decid- ever, the company manages reputational risks only infor-
ing whether to accept a given risk or to take actions to avoid mally – and unevenly – at the local and product levels. Its
or mitigate it. leaders consider reputational risk only when they make
major decisions such as those involving acquisitions. (The
company’s due-diligence process includes the evaluation of
The Current State of Affairs problems that could affect reputation, including pending
Regulators, industry groups, consultants, and individual com- lawsuits, weak product-testing procedures, product-liability
panies have developed elaborate guidelines over the years concerns, and poor control systems for detecting management
for assessing and managing risks in a wide range of areas, fraud.) The risk management VP says that reputational risk
from commodity prices to control systems to supply chains is not included in the long list of risks for which he is respon-
to political instability to natural disasters. However, in the sible. Then who is responsible? The CEO, the vice president

“ It takes many good deeds to build a good reputation,

and only one bad one to lose it.

” – Benjamin Franklin

absence of agreement on how to define and measure reputa- surmises, since that is who oversees the firm’s elaborate crisis-
tional risk, it has been ignored. response system and is ultimately responsible for dealing
Consider the 135-page framework for enterprise risk man- with any events that could damage the company’s reputa-
agement (ERM) proposed in 2004 by the Committee of tion. This pharmaceutical firm is not alone. Contingency
Sponsoring Organizations of the Treadway Commission plans for crisis management are as close as most large and
(COSO), a group of professional associations of U.S account- midsize companies come to reputational-risk management.
ants and financial executives that issues guidelines for inter- While such plans are important, it is a mistake to confuse
nal controls. Although the framework mentions virtually them with a capability for managing reputational risk. Know-
every other imaginable risk, it does not contain a single ref- ing first aid is not the same as protecting your health.
erence to reputational risk.
Nor does the Basel II international accord for regulating
capital requirements for large international banks. In de- Determinants of Reputational Risk
fining operational risk as “the risk of loss resulting from Three things determine the extent to which a company is ex-
inadequate or failed internal processes, people and systems posed to reputational risk. The first is whether its reputation
or from external events,” the Basel II framework, issued in exceeds its true character. The second is how much external
2004 and updated in 2005, specifically excludes strategic and beliefs and expectations change, which can widen or (less

Robert G. Eccles (reccles@[Link]) and Scott C. Newquist (snewquist@[Link]) are founders and managing directors
of Perception Partners, a firm based in West Palm Beach, Florida, that advises companies on corporate governance, corporate reporting, and
reputational risk. Roland Schatz ([Link]@[Link]) is the founder and CEO of the Media Tenor Institute for Media Analysis, a firm based
in Lugano, Switzerland, that helps organizations manage their reputations through strategic media intelligence.

106 Harvard Business Review | February 2007 | [Link]

 BP’s Sinking Image
Media coverage plays a large role in determining plosion at BP’s Texas City refinery, alleged tax evasion
a company’s reputation. The changing mix of positive in Russia, and job cuts in Europe took their toll in
and negative stories mentioning BP in the leading 2005, when positive and negative coverage were
British, German, and U.S. media from January 2003 roughly equal. Events in 2006 – especially an oil leak
through September 2006 shows how a series of at the Prudhoe Bay field in Alaska due to pipeline cor-
events hurt the oil giant’s reputation. During 2003 rosion, and a subsequent cut in production – caused
and 2004, the ratio of positive to negative stories the number of stories mentioning BP to soar and the
was about two to one. However, stories about an ex- mix to become more negative than positive.

100%
Job cuts in Europe POSITIVE
10/29/05 STORIES

Russian tax-evasion case settled

8/11/05 Prudhoe Bay
80% pipeline corrosion
Investigators of Texas City refinery revealed,
explosion release findings production cutback
8/18/05 8/7/06
Texas City refinery explosion Allegations of
60% 3/23/05 propane market
manipulation NO CLEAR
Allegations of tax evasion in Russia 6/29/06 RATING
11/12/04
Prudhoe
40%
Bay
leak
3/2/06

20%
NEGATIVE
STORIES

0
J F M A M J J A S O N D J F M A M J J A S O N D J F M A M J J A S O N D J F M A M J J A S
2003 2004 2005 2006

Dates are when first reports of events

YEL MAG CYAN BLACK

appeared in leading media.

likely) narrow this gap. The third is the quality of internal co- Reputation is distinct from the actual character or behav-
ordination, which also can affect the gap. ior of the company and may be better or worse. When the
Reputation-reality gap. Effectively managing reputational reputation of a company is more positive than its underlying
risk begins with recognizing that reputation is a matter of per- reality, this gap poses a substantial risk. Eventually, the fail-
ception. A company’s overall reputation is a function of its ure of a firm to live up to its billing will be revealed, and its
reputation among its various stakeholders (investors, cus- reputation will decline until it more closely matches the re-
tomers, suppliers, employees, regulators, politicians, non- ality. BP appears to be learning this the hard way. The energy
governmental organizations, the communities in which the giant has striven to portray itself as a responsible corporation
firm operates) in specific categories (product quality, corpo- that cares about the environment. Its efforts have included
rate governance, employee relations, customer service, intel- its extensive “Beyond Petroleum” advertising campaign and
lectual capital, financial performance, handling of environ- a multibillion-dollar initiative to expand its alternative-
mental and social issues). A strong positive reputation among energy business. But several major events in the past two
stakeholders across multiple categories will result in a strong years are now causing the public to question whether BP is
positive reputation for the company overall. truly so exceptional. (See the exhibit “BP’s Sinking Image.”)

[Link] | February 2007 | Harvard Business Review 107

Reputation and Its Risks

One was the explosion and fire at its Texas City refinery in Changing beliefs and expectations. The changing beliefs
March 2005 that killed 15 people and injured scores of oth- and expectations of stakeholders are another major deter-
ers. Another was the leak in a corroded pipeline at its Prud- minant of reputational risk. When expectations are shifting
hoe Bay oil field in Alaska that occurred a year later and and the company’s character stays the same, the reputation-
forced the company to slash production in August 2006. BP reality gap widens and risks increase.
has blamed the refinery disaster on lax operating practices, There are numerous examples of once-acceptable prac-
but federal investigators have alleged that cost cutting con- tices that stakeholders no longer consider to be satisfactory
tributed as well. Employee allegations and company reports or ethical. Until the 1990s, hostile takeovers in Japan were al-
suggest that the root cause of the Prudhoe Bay problem may most unheard of–but that was partly due to the cross-holding
have been inadequate maintenance and inspection practices of shares among the elite groups of companies known as kei-
and management’s failure to heed warnings of potential cor- retsu, a practice that undermined the power of other share-
rosion problems. As media coverage reflects, these events and holders. With the weakening of the keiretsu structure during
others have damaged BP’s reputation. the past ten to 15 years, shareholder rights and takeovers
To bridge reputation-reality gaps, a company must either have been on the rise. In the United States, once-acceptable
improve its ability to meet expectations or reduce expecta- practices now considered improper include brokerage firms
tions by promising less. The problem is, managers may resort using their research functions to sell investment-banking

“ Character is like a tree and reputation like its shadow. The shadow
is what we think of it; the tree is the real thing.

” – Abraham Lincoln

to short-term manipulations. For example, reputation-reality deals; insurance underwriters’ incentive payments to bro-
gaps concerning financial performance often result in ac- kers, which caused brokers to price and structure coverage to
counting fraud and (ultimately) restatements of results. Com- serve underwriters’ interests rather than customers’; the ap-
puter Associates, Enron, Rite Aid, Tyco, WorldCom, and pointment of CEOs’ friends to boards as “independent direc-
Xerox are some of the well-known companies that have tors”; earnings guidance; and smoothing of earnings.
fallen into this trap in recent years. Sometimes norms evolve over time, as did the now wide-
Of course, organizations that actually meet the expecta- spread expectation in most developed countries that com-
tions of their various stakeholders may not get full credit for panies should pollute minimally (if at all). A change in the
doing so. This often occurs when a company’s reputation has behavior or policies of a leading company can cause stake-
been significantly damaged by unfair attacks from special in- holders’ expectations to shift quite rapidly, which can imperil
terest groups or inaccurate reporting by the media. It also the reputations of firms that adhere to old standards. For ex-
can happen when a company has made genuine strides in ad- ample, the “ecomagination” initiative launched by General
dressing a problem that has hurt its reputation but can’t con- Electric in 2005 has the potential to raise the bar for other
vince stakeholders that its progress is real. For example, Chry- companies. It committed GE to doubling its R&D investment
sler, Ford, and General Motors improved their cars so much in developing cleaner technologies, doubling the revenue
that the quality gap between them and the vehicles made by from products and services that have significant and measur-
Japanese companies had largely closed by 2001. Yet, much to able environmental benefits, and reducing GE’s own green-
the frustration of the Big Three, consumers remain skeptical. house emissions.
Undeserved poor or mediocre reputations can be mad- Of course, different stakeholders’ expectations can diverge
dening. The temptation is to respond to them with resigna- dramatically, which makes the task of determining accept-
tion and conclude: “No matter what we do, people won’t like able norms especially difficult. When GlaxoSmithKline pio-
us, so why bother?” The reason executives should bother – neered the development of anti-retroviral drugs to combat
through redoubled efforts to improve reporting and com- AIDS, its reputation for conducting cutting-edge research
munications – is that their fiduciary obligation to close such and product development was reinforced and shareholders
reputation-reality gaps is as great as their obligation to im- were pleased. They were initially on board when GSK led
prove real performance. Both things drive value creation for a group of pharmaceutical companies in suing the South
shareholders. African government after it passed legislation in 1997 allow-

108 Harvard Business Review | February 2007 | [Link]

ing the country to import less expensive,
generic versions of AIDS drugs covered
by GSK patents. But in 2001, GSK share-
holders did an about-face in reaction to
an intensifying campaign waged by NGOs
and to the trial proceedings, which made
GSK and the other drug companies look
greedy and immoral. With its reputation
plunging, GSK relented and granted a
South African company a free license to
manufacture generic versions of its AIDS
drugs–but the damage was already done.
Sometimes, particular events can
cause latent concerns to burst to the sur-
face. One example would be all the ques-
tions about whether Merck had fully
disclosed the potential of its painkiller
Vioxx to cause heart attacks and strokes.
Merck is embroiled in thousands of law-
suits over the arthritis drug, which it
pulled from the market in 2004. The con-
troversy has raised patients’ and doctors’
expectations that drug companies should
disclose more detailed results and analy-
ses of clinical trials, as well as experience
in the market after drugs have received
regulatory approval.
When such crises strike, companies
complain that they have been found
guilty (in the courts or in the press) be-
cause the rules have changed. But all too often, it’s their own tion bonuses for senior managers and a big payment to a
fault: They either ignored signs that stakeholders’ beliefs and trust fund designed to protect executive pensions in the
expectations were changing or denied their validity. event of bankruptcy. However, the company didn’t tell the
In addition, organizations sometimes underestimate how unions. Furious when they found out, the unions revisited
much attitudes can vary by region or country. For example, the concessions package they had approved. The controversy
Monsanto, a developer of genetically modified plants, was cost CEO Donald J. Carty his job.
badly burned by its failure to anticipate Europeans’ deep con- Poor internal coordination also inhibits a company’s abil-

YEL MAG CYAN BLACK

cerns about genetically modified foods. ity to identify changing beliefs and expectations. In virtually
Weak internal coordination. Another major source of rep- all well-run organizations, individual functional groups not
utational risk is poor coordination of the decisions made by only have their fingers on the pulses of various stakehold-
different business units and functions. If one group creates ers but are also actively trying to manage their expectations.
expectations that another group fails to meet, the company’s Investor Relations (with varying degrees of input from the
reputation can suffer. A classic example is the marketing de- CFO and the CEO) attempts to ascertain and influence the ex-
partment of a software company that launches a large adver- pectations of analysts and investors; Marketing surveys cus-
tising campaign for a new product before developers have tomers; Advertising buys ads that shape expectations; HR
identified and ironed out all the bugs: The company is forced surveys employees; Corporate Communications monitors
to choose between selling a flawed product and introducing the media and conveys the company’s messages; Corporate
it later than promised. Social Responsibility engages with NGOs; and Corporate Af-
The timing of unrelated decisions also can put a com- fairs monitors new and pending laws and regulations. All of
pany’s reputation at risk, especially if it causes a stakeholder these actions are important to understanding and managing
group to jump to a negative conclusion. This happened to reputational risks. But more often than not, these groups do
American Airlines in 2003, when it was trying to stave off a bad job of sharing information or coordinating their plans.
bankruptcy. At the same time that it was negotiating a major Coordination is often poor because the CEO has not as-
reduction in wages with its unions, its board approved reten- signed this responsibility to a specific person. When 269

[Link] | February 2007 | Harvard Business Review 109

Reputation and Its Risks

executives were asked in 2005 by the Economist Intelligence ing volume and prominence of coverage, topics of interest,
Unit who at their companies had “major responsibility” for and whether the view is positive or negative.
managing reputational risk, 84% responded,“The CEO.” This Establishing a positive reputation through the media de-
means that nobody is really overseeing the coordination pro- pends on several factors or practices, according to research
cess. Yes, the CEO is the person ultimately responsible for by the Media Tenor Institute for Media Analysis (founded by
reputational risk, since he or she is ultimately responsible coauthor Roland Schatz) in Lugano, Switzerland.
for everything. But the fact of the matter is, the CEO does not First, the company has to land and remain on the public’s
have the time to manage the ongoing process of coordinat- radar screen, which involves staying above what we call the
ing all the activities that affect reputational risk. “awareness threshold”: a minimum number of stories men-
tioning or featuring the company in the leading media. This
volume, which must be continual, varies somewhat from
Managing Reputational Risk company to company, depending on industry and country
Effectively managing reputational risk involves five steps: as- but not on company size.
sessing your company’s reputation among stakeholders, eval- Second, a positive reputation requires that at least 20% of
uating your company’s real character, closing reputation- the stories in the leading media be positive, no more than
reality gaps, monitoring changing beliefs and expectations, 10% negative, and the rest neutral. When coverage is above
and putting a senior executive below the CEO in charge. the awareness threshold and is positive overall, the com-
Assess reputation. Since reputation is perception, it is per- pany’s reputation benefits from individual positive stories
ception that must be measured. This argues for the assess- and is less susceptible to being damaged when negative sto-
ment of reputation in multiple areas, in ways that are contex- ries appear. If coverage is above the awareness threshold but
tual, objective, and, if possible, quantitative. Three questions the majority of stories are negative, a company will not ben-
need to be addressed: What is the company’s reputation in efit from individual positive stories, and bad news will rein-
each area (product quality, financial performance, and so force its negative reputation. All companies – large or small –
on)? Why? How do these reputations compare with those of should care about staying above their awareness threshold.
the firm’s peers? Even if a small company has a very strong reputation among
Various techniques exist for evaluating a company’s repu- a small group of core investors or customers, it runs a high risk
tation. They include media analysis, surveys of stakeholders of suffering considerable damage to its reputation if its media
(customers, employees, investors, NGOs) and industry exec- coverage is below the awareness threshold when a crisis hits.
utives, focus groups, and public opinion polls. Although all A company’s reputation is also vulnerable if the media are
are useful, a detailed and structured analysis of what the focused on just a few topics, such as earnings and the person-
media are saying is especially important because the media ality of the CEO. Even if the coverage of these topics is ex-
shape the perceptions and expectations of all stakeholders. tremely favorable, a negative event outside these areas will
Today, many companies hire clipping services to gather have a much larger negative impact than it would have if the
stories about them. Text- and speech-recognition technolo- firm had enjoyed broader positive coverage.
gies enable these services to scan a wide range of outlets, in- Third, managers can influence the mix of positive, negative,
cluding newspapers, magazines, TV, radio, and blogs. They and neutral stories by striving to optimize the company’s
can provide information on such things as the total number “share of voice”: the percentage of leading-media stories
of stories, the number per topic, and the source and author mentioning the firm that quote someone from the organiza-
of each story. While useful in offering a real-time sample of tion or cite data it has provided. Media Tenor’s research sug-
media coverage, these services are not always accurate in as- gests that a company needs to have at least a 35% share of
sessing whether a story about a company is positive, nega- voice in order to keep the proportion of negative stories to a
tive, or neutral, because of the limits of the computer algo- minimum in normal times. Strong relationships and credibil-
rithms that they employ. They also tend to miss stories that ity with the press are crucial to attaining a large share of voice
cite a company but do not mention it in the headline or first and are especially important during a crisis, when a company
few sentences. really needs to communicate its point of view. In such times,
Therefore, the old tool of clipping services needs to be management’s share of voice needs to be at least 50% to en-
supplemented with strategic media intelligence. This new sure that critics of the company don’t prevail. Merck’s travails
tool not only analyzes every line in a story but also places the after the problems with Vioxx illustrate the consequences of
coverage of a company within the context of all the stories in a company inadequately managing its position in the media.
the leading media (those that set the tone for the coverage of (See the exhibit “Merck: The Perils of a Low Profile.”)
topics, companies, and people in individual countries). Since Evaluate reality. Next, the company must objectively eval-
the reputation of a company is a function of others’ reputa- uate its ability to meet the performance expectations of
tions in its industry and the relative reputation of the indus- stakeholders. Gauging the organization’s true character is
try overall, having the complete context is essential for assess- difficult for three reasons: First, managers–business unit and

110 Harvard Business Review | February 2007 | [Link]

 Merck: The Perils of a Low Profile
Merck was ill prepared to defend its reputation when the Vioxx cited data provided by it, meaning Merck didn’t have the “share
crisis hit. In the 33 months prior to Merck’s withdrawal of the of voice” required to communicate its positions. After the an-
pain medication on September 30, 2004, the company had a low nouncement of Vioxx’s withdrawal, the average number of sto-
profile: There weren’t enough leading-media stories mentioning ries per month mentioning Merck more than tripled – but 60% of
it to keep it above the public’s “awareness threshold.” Although the stories that appeared through September 2006 were nega-
27% of the stories were positive, they were neutralized by the tive and only 13% positive. It will be difficult for Merck to rebuild
28% that were negative. In addition, before the recall, a woefully its reputation – especially since its share of voice has decreased
inadequate 7% of stories quoted someone from the company or to 5.5%.

140
Date of Vioxx announcement
September 30, 2004

POSITIVE
STORIES
120

NO CLEAR
RATING
100

Number NEGATIVE
of reports
STORIES Average number of
reports per month
80
after September 30, 2004: 77

60

AWARENESS THRESHOLD

40

Average number of

YEL MAG CYAN BLACK

reports per month
before September 30, 2004: 21
20

0 J F M A M J J A S O N D J F M A M J J A S O N D J F M A M J J A S O N D J F M A M J J A S O N D J F M A M J J A S
2002 2003 2004 2005 2006

functional heads as well as corporate executives–have a nat- they are set low in order to ensure that performance objec-
ural tendency to overestimate their organizations’ and their tives will be achieved, and other times they are set optimisti-
own capabilities. Second, executives tend to believe that cally high in an attempt to impress superiors or the market.
their company has a good reputation if there is no indication As is the case in assessing reputation, the more contextual,
that it is bad, when in fact the company has no reputation objective, and quantitative the approach to evaluating char-
in that area. Finally, expectations get managed: Sometimes acter, the better. Just as the reputation of a company must

[Link] | February 2007 | Harvard Business Review 111

 Reputation and Its Risks

One Drug Company’s Dashboard for Spotting Potential Risks

A Europe-based pharmaceutical com- are ahead of plan, the Central Nervous the projected shortfall – such as special
pany uses this dashboard to track vari- System & Pain division is projecting that incentive programs for the sales force
ances in performance that can lead to its revenues for the full year will fall short, or prescribing physicians – would create
risky behavior. (The names of divisions mainly because of the Ibellance brand’s unacceptable reputational risk for the
and brands have been changed and the projected performance. At this point, cor- company. And if Ibellance’s performance
data have been scaled to protect propri- porate executives should meet with the dramatically improves during the rest of
etary information.) Although the reve- division’s managers to ensure that none the year, they would be wise to investi-
nues of the company’s U.S. business of the unit’s planned actions to address gate again.

Variances from Budget on Revenue Projections, Year-to-Date

Total revenues of the U.S. business
are ahead of budget by $350.1M

Digestive Allergy & Inflammation Central Nervous Oncology Other Other

Division Respiratory Division System & Pain Division Pharmaceuticals Businesses
$46.17M Division $50.67M Division $190.84M $20.34M −$3.78M
$119.09M −$73.23M

Deipand brand Prelfin brand Elmaxol brand Thivas brand Gaprivert brand
$50.40M −$32.58M $30.67M $7.90M $154.45M

Mangeium brand Transbis brand Osisequn brand Itrarig brand Lobane brand
−$7.04M $93.38M −$5.43M −$15.09M $16.96M

All other marketed Ectosyceium brand Throac brand Hispecart brand Oprav products
digestive products $0M $16.64M −$12.97M −$0.01M
$2.81M

Prasess brand Aphervir brand Ibellance brand Smoodent brand

$0M −$3.10M −$43.44M $13.92M

Ectonab brand Othuctage brand Othedicize brand Grispec brand

$62.81M $15.19M −$3.74M $2.89M

Brevdorp brand Oloxic brand All other marketed All other oncology
−$1.88M $0M CNS & Pain products products
−$5.89M $2.63M
Analysis conducted with Empower visualization
software, by Metapraxis.

All other marketed All other marketed Ahead of budget

A&R products inflammation
On or close to budget
−$2.64M products (variance <5%)
−$3.3M
Below budget (variance >5%)

112 Harvard Business Review | February 2007 | [Link]

be assessed relative to competitors, so must its reality. For Company’s Dashboard for Spotting Potential Risks” for an
example, performance-improvement targets based only on example of a simple but effective use of visualization soft-
a company’s results for the previous year are meaningless if ware to highlight whether business units and products are on
competitors are performing at a much higher level. The im- track to meet year-end goals.)
portance of benchmarking financial and stock performance Close gaps. When a company’s character exceeds its repu-
and processes against peers’ and those of companies re- tation, the gap can be closed with a more effective investor
garded as “best in class” is hardly a revelation. However, the relations and corporate communications program that em-
degree of sophistication and detail as well as the accuracy or ploys the principles of strategic media intelligence discussed
reliability of benchmarking data can vary enormously. The above. If a reputation is unjustifiably positive, the company
reasons include transcription errors (a big problem when a must either improve its capabilities, behavior, and perfor-
large amount of data in paper documents has to be manually mance or moderate stakeholders’ perceptions. Of course, few
entered into electronic spreadsheets), for instance, and the companies would choose the latter if there were any way to
inability to determine whether the way competitors report accomplish the former. If, however, the gap is large, the time
information in an area is consistent. One company might in- required to close it is long, and the damage if stakeholders
clude customers’ purchases of extended warranties in its rev- recognize the reality is likely to be great, then management
enues, while another might not. should seriously consider lowering expectations – although
Some new tools should help address these issues. One of this obviously needs to be done in careful, measured ways.
the most noteworthy is Extensible Business Reporting Lan- Monitor changing beliefs and expectations. Understand-
guage (XBRL). A version of the Internet standards technol- ing exactly how beliefs and expectations are evolving is not
ogy Extensible Markup Language (XML), XBRL allows each easy, but there are ways to develop a picture over time. For
piece of information in a financial statement to be electron- instance, regular surveys of employees, customers, and other
ically tagged so that it can be quickly and cheaply pulled into stakeholders can reveal whether their priorities are chang-
analytical software. These tags are contained in dictionaries, ing. While most well-run companies conduct such surveys,
or “taxonomies,” based on sets of standards such as the U.S. few take the additional step of considering whether the data
generally accepted accounting principles. XBRL-formatted fi- suggest that a gap between reputation and reality is materi-
nancial statements are already available from companies alizing or widening. Similarly, periodic surveys of experts in
such as EDGAR Online, but these early offerings have limi- different fields can identify political, demographic, and social
tations. Taxonomies for specific industries must be devel- trends that could affect the reputation-reality gap.“Open re-
oped; software for downloading and analyzing XBRL data is sponse” questions can be used to elicit new issues of impor-
still at an early stage; and EDGAR Online’s offering includes tance–and thus new expectations–that other questions might
European companies only if their shares are listed on a U.S. miss. It is generally useful to supplement these surveys with
exchange (although an XBRL taxonomy does exist for inter- focus groups and in-depth interviews to develop a deeper un-
national financial reporting standards, used by all members derstanding of the causes and possible consequences of trends.
of the European Union and a number of other countries). Influential NGOs that could make the company a target
Christopher Cox, the chairman of the Securities and Ex- are one group of stakeholders that should be monitored.
change Commission, is determined to address such limita- These include environmental activists; groups concerned
tions and accelerate the widespread adoption of XBRL. To- about wages, working conditions, and labor practices; con-

YEL MAG CYAN BLACK

ward that end, he announced in September 2006 that the sumers’ rights groups; globalization foes; and animals’ rights
SEC will invest $54 million in an interactive data system groups. Many executives are skeptical about whether such
based on XBRL, which “will represent a quantum leap over organizations are genuinely interested in working collabora-
existing disclosure technologies.” (For more detail, see the tively with companies to achieve change for the public good.
HBR List item “Here Comes XBRL” in this issue.) But NGOs are a fact of life and must be engaged. Interviews
Another valuable new tool for managing reputational risk with them can also be a good way of identifying issues that
is visualization software, which uses colors, shapes, and dia- may not yet have appeared on the company’s radar screen.
grams to communicate the key points in financial and oper- Finally, companies need to understand how the media
ating data. These displays are a big improvement over the shape the public’s beliefs and expectations. Dramatic changes
spreadsheets now widely used, which often make it difficult in the amount of coverage influence how fast and to what ex-
for even the most financially sophisticated executives to spot tent beliefs and expectations change. The large volume and
important anomalies and trends. Because it takes so much prominent display of stories on the backdating of stock op-
time to make sense of spreadsheets, executives tend to focus tions in recent months is one example of how the media can
on the largest business units even though the greatest risks help set the agenda. The sharp drop in stories about insur-
to reputation may reside in smaller ones–such as a struggling ance brokers’ getting incentive payments from underwriters
foreign subsidiary that has begun to employ questionable illustrates how the media can help relegate a hot topic to the
means to meet budget targets. (See the exhibit “One Drug back burner.

[Link] | February 2007 | Harvard Business Review 113

Reputation and Its Risks

A Framework for Managing Reputational Risk

Understanding the factors that determine reputational risk enables a

company to take actions to address them.

DETERMINANTS OF Reputation-reality gap Changing beliefs Weak internal

REPUTATIONAL RISK and expectations coordination

WAYS TO MANAGE Objectively assess Assess and accept impact Explicitly focus on
REPUTATIONAL RISK reputation versus reality of changing expectations reputational risk

Examine the gap between the Know that stakeholders’ changing Recognize that this is a distinct
company’s reputation and actual expectations will affect reputa- kind of risk and manage it in a
performance; make necessary tion even if they seem unreason- proactive and coordinated man-
improvements. able at the time. ner. Assign one person the task
of managing reputational risk.

Strong and sustainable reputation

Put one person in charge. Assessing reputation, evaluating most well-managed companies, many of the elements are al-
reality, identifying and closing gaps, and monitoring chang- ready in place in disparate parts of the organization. The ad-
ing beliefs and expectations will not happen automatically. ditional costs of installing and using the new tools described
The CEO has to give one person responsibility for making above to identify risks and design responses are in the low to
these things happen. Obvious candidates are the COO, the high six figures, depending on the size and complexity of the
CFO, and the heads of risk management, strategic planning, company. This is a modest expense compared with the value
and internal audit. They have the credibility and control at stake for many companies.
some of the resources necessary to do the job. In general, So the primary challenge is focus: recognizing that repu-
those whose existing responsibilities pose potential conflicts tational risk is a distinct category of risk and giving one per-
probably shouldn’t be chosen. People holding top “spin”jobs, son unambiguous responsibility for managing it. This person
such as the heads of marketing and corporate communica- can then identify all the parts of the organization whose ac-
tions, fall into this category. So does the general counsel, tivities can affect or pose risks to its overall reputation and
whose job of defending the company means his relationship enhance the coordination among its functions and units. The
with stakeholders is often adversarial and whose typical re- improvements in decision making will undoubtedly result in
sponse to media inquiries is “no comment.” a better-run company overall.
The chosen executive should periodically report to top Senior executives tend to be optimists and cheerleaders.
management and the board on what the key reputational Their natural inclination is to believe the praise heaped on
risks are and how they are being managed. It is up to the their companies and to discount the criticism. But looking at
CEO or the board to decide whether the risks are acceptable the world and one’s organization through rose-tinted glasses
and, if not, what actions should be taken. In addition, top man- is an abdication of responsibility. Being tough-minded about
agement and the board should periodically review the risk- both will enable a company to build a strong reputation that
management process and make suggestions for improving it. it deserves.
•••

Managing reputational risk isn’t an extraordinarily expen- Reprint R0702F

sive undertaking that will require years to implement. At To order, see page 158.

114 Harvard Business Review | February 2007 | [Link]

Copyright 2007 Harvard Business Publishing. All Rights Reserved. Additional restrictions
may apply including the use of this content as assigned course material. Please consult your
institution's librarian about any restrictions that might apply under the license with your
institution. For more information and teaching resources from Harvard Business Publishing
including Harvard Business School Cases, eLearning products, and business simulations
please visit [Link].