The Real Lesson

of New Coke:
The Value of Focus Groups for Predicting the Effects of Social Influence
The embarrassing failure of Coca-Cola's attempt to change the flavor of itsflagshipbrand has hecome a textbook case of how ma
ket research can fail. The lesson usually drawn is that Coke's researchers asked respondents the wrong questions. However, a
careful examination of the events surrounding the reformulation attempt suggests an alternative explanation: that the error res
ed from the standard market research practice of considering focus groups to be only a form of preliminary research and not app
ciating their unique ability to predict the effects of social influence.

By Robert M. Schindler
Pepsi began communicating these find- ferred the new formula over the old for-
n April 1985, the man-
ings to consumers through "Pepsi Chal- mula hy 61 % to 39%,
agement of Coca-Cola
lenge" television ads .showing taste tests Well aware of the importance of the
Co. announced its de-
where Coca-Cola drinkers expressed pref- reformulation decision. Coke's manage-
cision to change the fla-
erences for a cola which was then revealed ment made sure that the taste test results
vor of the cotnpany's
to be Pepsi, This campaign contributed to were checked and coiToborated in every
flagship brand. The
Coca-Cola's slow, but steady decline of major market in the country. Overall,
events that followed from this decision, as
market share in the soft-drink category. Coke's market researeh on the reformula-
well as the faetors which led up to it, have
This erosion was most apparent in food- tion was one of the most exhaustive mar-
been reviewed, discussed, and extensively
store sales, which reflect consumer prefer- ket research projects in history; It cost $4
analyzed in the popular press, the trade
ences more directly than do vending- million and included interviews with
press, and in marketing textbooks.
tiiachine or fountain sales. By 1977, Pepsi almost 200,000 consumers. After the deci-
Two books and at least two marketing sion to reformulate Coke was made, Coca-
had actually pulled ahead of Coke in food-
cases have been written on the events sur- Cola chairman, Roberto Goizueta, termed
store market share.
rounding the flavor change decision. Also, the decision, "one of the easiest we have
a well-known, but somewhat older Har- Although publicly expressing a lack of
ever made," according to Hale N, Tongren
vard Business School marketing case deals concern about the Pepsi Challenge adver-
in his book. Cases in Consumer Behavior.
with some of the key events which led up tising, Coca-Cola's managetnent privately
On April 23, 1985, Coke announced the
to the decision. Despite the extent of this was quite worried because blind taste tests
reformulation with a grand flourish, slag-
attention, more can be learned from this by the company's own market research
ing a multicity satellite press conference in
dramatic pieee of marketing history. department had confirmed Pepsi's claims.
New York, Atlanta, Chicago, Houston,
Secretly, Coke's management began
Los Angeles, and Toronto. The next day, a
researching the possibility of reformulat-
The Attempt to Reformulate ing Coca-Cola to respond to the apparent
front-page article in The New York Times
reported: "The Coca-Cola Company said
Coca-Cola changes that had occurred in consumer
yesterday that il had scrapped the formula
tastes. By 1984, researchers had arrived at
for the world's best-selling soft drink. The
a new formula for Coke whieh, in blind

T
he 87-year old rivalry between Coca- recipe, concocted 99 years ago, has been
Cola, the traditional market leader, taste tests, beat Pepsi by as much as six to
piaeed in the vault at the Trust Company
and Pepsi Cola, the perennial runner eight percentage points. In addition to
of Georgia Bank, never to be used again,
up, took an unexpected turn in the mid- beating Pepsi, cola drinkers chose this new
said Roberto C. Goizueta, chairman of
1970s. Pepsi's consumer research discov- formula over the old Coke formula by
Coca-Cola. "We have a new formula for
ered in blind taste tests that a majority of 55% to 45% in blind taste tests and loyal
Coke," he added."
consumers preferred the taste of Pepsi Coke drinkers chose it over the old Coke
formula by 53% to 47%, In taste tests In addition to the extensive publicity.
over thai of Coke. In fact, even a majority
where the drinks were identified as "new Coke announced that the new Coke would
of loyal Coke drinkers reported preferring
Coke" and "old Coke," cola drinkers pre- come in a new ean, with updated red and
Pepsi in the tests.

22 Decetnber1992
silver graphics replacing the traditional red Goizueta explained that the company had back the old formula. Coke's president.
and white look. Clearly. Coke had decided adopted a "megabrand strategy" in which Donald Keough told Time magazine,
to make sure that consumers would be some promotion expenditures were direct- "...you cannot measure it any more than
aware that Coca-Cola's flavor was being ed at enhancing the image of the compa- you can measure love, pride, or patrio-
changed. ny's entire line of cola drinks. tism,"
The initial reactions of most consumers During the following years, the market However, most observers did not
appeared to be positive. Many bottlers share of new Coke continued to decline. attribute the failure of Coke's research in
reported that sales of new Coke were By 1990. new Coke's share had dropped this instance to an intrinsic limitation of
greater than expected and. during the first to .6%, In April 1990, the company began the capabilities of marketing research.
few weeks after the new Coke introduc- test marketing new Coke under the name Rather, they judged that the research was
tion, the company's weekly survey of 900 •'Coke II." It was advertised as having conducted or interpreted incorrectly.
respondents showed consumers preferring "real cola taste" with "the sweetness of Although some have argued that Coke's
new Coke over old Coke by a margin of Pepsi," according to the Wall Street Jour- research error was to overgeneralize from
53% to 47%. nal. Although Coke's management has inexact taste test results, the vast majority
However, during this period, there was claimed that these tests were successful, of people who have publicly voiced an
also intense media coverage of those con- they appeared to have developed a consid- opinion concerning where Coke's research
sumers who did not like the new Coke and erable degree of caution on the question of efforts went wrong espouse what could be
were angry about the change. In a number supporting Coke TT with a national rollout. called the "wrong-question explanation."
of cities, old Coke loyalists sponsored In this view, the reason that Coke's
protest rallies and boycotts and received marketing research did not detect the con-
widespread media attention. sumer outcry which resulted from the
By June, it was becoming apparent to reformulation was that they did not make
Coke's management that consumer dissat- it clear to the taste-test respondents that if
isfaction with the reformulation was most people chose the new Coke flavor,
increasing. The stream of angry letters and then the traditional Coke flavor would no
phone calls was becoming a flood, and longer be available. In other words, rather
weekly tracking surveys confirmed that
marketers have used than ask. "Which flavor do you like bet-
consumers were becoming increasingly these events to disparage ter," consumers should have been asked a
negative about the change. In a survey more relevant question. This question
conducted during the first week of July the discipline of , might have been something like. "How
only 30% of consumers interviewed marketing research I would you feel if we discarded Coca-
reported preferring the new Coke to the Cola's current taste and replaced it with
old. this new taste?"
On July 10. the company announced its In the years since the introduction of
decision to respond to public pressure and new Coke and the reintroduction of Coca-
bring back the old Coke formula. It would Cola Classic, the wrong-question explana-
be available in the form of a product with The Wrong-Question tion is the one that has been most often
the name "Coca-Cola Classic," and was repeated. It appears to have become the
intended as a flanker brand to satisfy those
Explanation standard explanation for this highly publi-
consumers who wanted the original taste cized failure of market research.
as an option. The reformulated soft drink
was to be known simply as "Coke" and
would remain the flagship brand. T he widespread awareness of these
events has led many observers to
search for an explanation. The par-
An Alternative Explanation
However, sales of new Coke eroded
rapidly. In August, sales of Classic began
to exceed those of new Coke, and by the
end of September, Classic had a 70%
ticular question which most frequently has
arisen is why Coke's extensive market
research was unable to provide manage-
ment with better guidance in the reformu-
S ince the intense publicity has died
down, some further details of the
research behind the new Coke deci-
sion have come to light. In particular, it is
share of the combined volume of the two lation decision. now known that Coca-Cola's market
products. Over the next few months, large Anyone who is or has been a marketing research department did indeed ask the
fountain accounts, such as McDonald's, research practitioner will not find it sur- right question. Considerable attention was
began switching back to the old formula. prising that some marketers have used devoted to testing consumer reactions to
In 1986. Classic outsold new Coke by these events to disparage the discipline of the idea of changing Coke's flavor.
more than 8 to 1. despite the promotion of marketing research. For example, in Mar- These tests and their results are
new Coke with over $48 million of top- keting News, one ad-agency manager described in Thomas Oliver's book. The
rated television advertising. Although attributed Coke's mistake to "executives Real Coke. The Real Story: "In 1982 the
Coke's advantage over Pepsi in the sug- who solely rely on statistics." Coca-Cola Company conducted 2,000
ared cola category had decreased slightly When announcing the re introduction of interviews in 10 major markets to investi-
compared to 1984. Coke's advantage over old Coke in July 1985, the top people at gate further the public's willingness to
Pepsi in the overall market had increased, Coke suggested that research is not capa- accept a different Coke,
mostly due to the continued success of ble of measuring the types of consumer "Consumers were shown storyboards,
Diet Coke and tbe introduction of Cherry feelings that resulted from the attempted comic-striplike mock commercials. One
Coke, In response to questions about reformulation. When describing the emo- board said Coke had added a new ingredi-
which was now the flagship brand, tional outpouring which led them to bring ent and it tasted smoother, while another

December 1992 2 3
 said the same about Pepsi. (To compensate Coke's only deviation from this standard results of the focus groups and the survey
for the impact of first impressions, 1,000 sequence is that the quantitative survey of of individuals by trusting only the survey.
people were shown the Coke statement individuals appears to have been done As it turned out. one can see that both pro-
first and the other 1,000 saw the Pepsi before rather than after the focus groups. cedures had provided important informa-
announcement first.) Then Coca-Cola But this is a minor point. What we see here tion. When new Coke was first introduced.
asked consumers a long series of questions are very normal market research proce- people made individual decisions on it.
about what their reactions to such a dures. Coke's researchers started out asking and most at least acquieseed to the change.
change would be. Would you be upset? the right questions, and in the right way. This is what was predicted by the indi-
Would you try the new drink? Would you The results of the foeus-group phase and vidual interviews which indicated that
switch brands immediately? the survey conflicted. The researchers only \{)%-\2% of consumers would be
"'We estimated from the response that trusted the survey, whieh comprised a large upset. But over time, as the majority of the
10%-12%i of exclusive Coke drinkers number of interviews spread over a wide, population bad the opportunity to be stim-
would be upset, and that half of those and presumably representative, area. ulated by media reports and other social
would get over it, but half wouldn't,' said interactions with angry Coke loyalists.
|Roy] Stout [director of marketing most changed their minds. This is what
research I. was predicted by the focus groups. Given
"...While the interviews pointed to peo- the IO%-12% figure from the quantitative
ple's willingness to try a new Coke, survey, a typical eight- to 12-member
Zyman [vice-president of marketing] and - he focus focus group is likely to have at least one
Stout discovered through other tests that group is a unique source of infor- angry loyalist as a member. The focus
many people just didn't believe anyone group results showed that, in this situation,
could or should tamper with the king of mation about how the consumer exposure to the views of angry Coke loy-
the colas. To hear debate on the issue, will respond in a situation where alists is likely to sway the others in the
Coca-Cola's research department used group to their position.
focus groups, a favorite marketing tool. there will be an awareness of the By July 1985, Coke executives had
"...Some of Stout's focus groups were views of other consumers sensed that this social interaction was a
shown a storyboard depleting a proposed major factor in causing their problems; it
commercial. One said Brand X soft drink was reported in Advertising Age that Coke
was going to be improved. 'Fine.' the A wave of focus groups usually officials were blaming the press for "fan-
group replied, and they were equally san- involves less than 200 respondents and is ning public discontent." Of course, by
guine about a proposed improvement in conducted with minimal attention to then it was too late. Coke had already
Pepsi, But when it came to changing Coke obtaining a representative sample. It is Ignored the research that told them how
for the better, the resounding response was standard market research practice to trust the market would respond to a flavor
NO. 'It was like saying you were going to survey research over focus groups, and change carried out in a public context.
make the flag prettier.' said Zyman. A this praetiee appears, even to the nonre- Moreover, by the summer of 1985,
similar response came in 1983 from a searcher (i.e.. the executive decision- Pepsi had also come to appreciate the role
group that included some exclusive Coke maker), to make good sense. of publicity in causing Coke's problems,
drinkers. This group agreed that An- Moreover, Coke's research did a pretty and then actually acted on that understand-
heuser-Busch could change Budweiser but good job of predicting consumer response, ing. As David Gilman, manager of public
in no way should the Coca-Cola Company at least initially. Recall that when the relations for Pepsi Cola International,
try to improve Coke." reformulation was first introduced, the described it in Public Relations Journal:
Oliver goes on to describe how these consumer response was favorable. But by "After new Coke was introduced here
results led Coke managers to continue to the end of May 1985, it had begun to in the States, Coca-Cola said that it was
support the work of their technical people change. It was this that Coke had not planning to announce the new formulation
to come up with a new flavor for Coke expected. overseas. But months went by, and they
that would beat Pepsi. By September The New York Times' report on Coke's never made the announcement to the for-
1984. product research had developed announcement of the reintroduction of old eign market. This didn't sit well with us,
sueh a flavor, and taste-testing market Coke began as follows: "When the Coca- so as a public service we conducted press
research began moving into high gear. Cola Company introduced a reformulated conferences in 18 different countries on
What Oliver describes sounds very version of the world's best-selling soft Aug. 20. In effect, we made Coke's
much like what would be considered stan- drink on April 23, it was well aware that it announcement for them."
dard market researeh proeedure for the might alienate some faithful Coke In this view, the real lesson to be
development of a new product or the mod- drinkers. The company, however, expect- learned from the new Coke affair is that
ification of an existing one. The ideal ed that alienation to fade. It was complete- the focus group technique is more than
would be to begin with focus-group test- ly unprepared for how it would spread and just a means of getting a quick and vivid
ing of the product concept. Then, a survey deepen in the two months following the look at consumer opinion. It is a unique
would be conducted, using individual debut of the new Coke." source of information about how the con-
interviews with a large representative sam- It is this change in eonsumer opinion, sumer will respond in a situation where
ple of consumers, to verify and quantify and only this change, that Coke's market there will be an awareness of the views of
the results of the focus groups. If the new research had failed to predict. other consumers.
product survived these concept tests, then With the benefit of hindsight, the locus Even if Coke's researchers had includ-
the testing of an actual product (or proto- of their research mistake becomes clear. It ed in all of their individual interviews and
type) would begin. was to respond to the conflict between the taste tests a question directly asking con-

24 December 1992 Marketing Research

sumers how they would respond if only Although the interpersonal interactions products and felt that it was devious and
tbe new Coke flavor were available, it is in a focus group setting cannot be expect- insulting to try to pass it off as a new
very unlikely that the results would have ed to be identical to those that will occur product. This invariably led to a change in
predicted the shift in opinion which even- under natural conditions, they can be the views of the other group members, and
tually occurred. Without information on expected to yield a far better indication of the consensus of most of the focus groups
how other consumers will respond, indi- the effects of group interaction than meth- was negative. Note that the difference
viduals have no means of predicting how ods where consumers are interviewed indi- between the individual and group results
their own feelings will change after being vidually. In individual interviews, whether in this ease was due to the ability of the
exposed to the responses of others. qualitative or questionnaire-based, there is focus group to bring to awareness a view-
Viewing the events surrounding Coke's no opportunity at all for individual opin- point which was probably not even con-
attempted reformulation in this light, one ions to be influenced by the views of other sidered during most of the individual
is less likely to fault Coca-Cola's manage- eonsumers. interviews.
ment. After reintroducing the old Coke, This distinctive strength of the focus Apparently, such also was the case with
these executives were subjected to unflat- new Coke. When interviewed individually
tering suggestions that they had failed to about the proposed refonnulation, the vast
appreciate the importance of Coca-Cola in majority of the respondents were not even
the American psyehe or that they had eon.scious that some others would express
absentmindedly neglected to ask con- feelings of loss and anger. However, the
sumers an obviously relevant question. -1 is the ability of the group setting to make them
Rather, with the perspective of some active awareness conscious of such views as they were for-
distance, it appears that they indeed appre- mulating their opinions made it possible
ciated the importance of consumer of the opinion for them to be influenced by this opinion.
response to the idea of reformulating Coke of influential others During the introduction of new Coke, it
and used very acceptable market researcb was largely the media rather than personal
procedures to try to predict consumer which is the key factor interactions which caused this conscious-
reaction to this idea. Moreover, in contrast ness, but the effect of this awareness on the
to many executives who devote resourees opinions of most consumers was the same.
to market research studies and then ignore
the results of these studies in favor of their group method is not important when indi- The Validity of Focus
own personal intuition, it appears that viduals can accurately anticipate the views
Goizueta and other top Coke executives of influential others and are actively con- Group Results
actually trusted and used their market scious of these views when they are being
research results.

T
interviewed. For example, suburban his new appreciation of the distinc-
It is a flaw in accepted market researcb homeowners are likely to be highly and tive capability of the focus group
practice, and in the understanding of con- accurately aware of how their neighbors method leads to the following practi-
sumer behavior that supports this practice, would react to a new line of exterior house cal advice. In new-products research,
which must bear at least some of the paints featuring iridescent colors. In such rather than have focus groups be the pre-
responsibility for the embarrassing failure eases, individual and group methods will liminary research and the individual inter-
of the Coca-Cola refonnulation. yield the same results and the distinctive views of a quantitative survey be the con-
strength of focus groups is not important. firming research, conduct preliminary
Predicting the Effects of However, when individually inter- research using both individual and group
methods. If the two forms of qualitative
viewed consumers either cannot or do not
Social Influence accurately anticipate the views of others, research agree, then the confirmatory
then the ability of a focus group to bring research ean proceed using the traditional

T he failure of market researchers to tbese views to awareness becomes an survey methods.

appreciate the important and unique important advantage. However, if the results of the prelimi-
information which focus groups pro- It is the active awareness of the opinion nary individual interviews and focus
vide may be one reason behind what of influential others which is the key fac- groups diverge, then the researcher needs
Edward Tauber has argued is the relatively tor here. Consider the case of medical doc- to consider how much awareness of the
poor validity of eoneept-testing research. tors being interviewed by a drug company views of influential others will exist in the
However, the recent rise in theoretical concerning a new product that dispensed marketing situation the research is being
work on qualitative methods has set the two types of birth-control pills which used to predict.
stage for a rethinking of the traditional needed to be taken together.- When inter- The researcher should carry out confir-
role of focus group methods. viewed individually, physicians indicated matory research using a procedure that
Consumer behaviorists have pointed a generally positive reaction to the new provides tbe respondent with a level of
out that focus groups and individual inter- produet. When interviewed in focu,s awareness of tbe views of others compara-
views each have distinctive strengths and groups, physicians initially expressed pos- ble to that which is expected to oecur dur-
weaknesses, and that each is appropriate itive reactions. ing tbe rollout of the marketing program in
for different researeh tasks. The opportu- However, at some point in the typical question. If the researcher expects that
nity for respondents to interact with others group, one physician would express a very very few consumers will be aware of the
makes focus groups the best method for negative view to the effect that he or she reactions of other consumers to the intro-
measuring the effects of interpersonal perceived this new product as nothing duction of the new or altered product, then
influence. more than a repackaging of two existing confirmatory research using individual

December 1992 2 5
interviews should yield the most accurate involves the alteration of an existing prod- careful attention should be given to the
prediction. uct, then even if the product is associated social context of the consumer decisions
But if the researcher expects that many with high visibility, importance, or diffi- that will determine the product's success
consumers will be aware of the reactions culty of decision, there are many cases or failure.
of others, then the focus group results are where management can choose to roll out Although focus groups are the best
more likely to be correct and should be the program with little or no consumer research method for revealing critical
confirmed with a research technique that awareness. social interaction effects, there is a need
gives respondents a realistic awareness of Coke's reformulation was almost cer- for methods which make possible the con-
the views of other consumers. tainly one of those eases. A gradual phase- firmation of the insights of focus group
At least three factors can be used to in of the new flavor with no announce- research. One approach would be the
estimate the level of avi'areness of the ment or other changes would very likely development of procedures for conducting
opinions of others: have gone unnoticed. Pepsi had been qui- group research on a large representative
etly reformulated from time to time, and sample of consumers.
•The visibility of product. Products some believe that Coke had. in f'act, For example, such "confirmatory focus
that are highly visible, such as automo- secretly modified its formula .several times groups" might be larger and less homoge-
biles, beer, clothing, and magazines dis- during its first 99 years of existence. Con- neous than traditional exploratory focus
played on the living-room coffee table, are sider also that researcher David Pierce groups, and recruitment could be guided
very likely to be associated with high found that, even after all the flavor-change by expectations concerning the types of
awareness of the views of other con- opinions consumers are likely to become
sumers. aware of after the product is introduced.
•The importance of the product. A second approach to the development
Consumers will tend to seek out tbe opin- of research techniques to verify focus
ions of influential others when making group results would be to experiment with
decisions about risky and otherwise onsideration individual interview procedures that make
important products such as medical ser- of the effects of social consumers aware of the full range of view-
vices, day-care centers, and banking ser- points expressed during the exploratory
vices. influence must become focus group pha.se and which are likely to
• T h e difficulty of the decision a standard part of the eome into eonsumer awareness after the
regarding the product. Even if the prod- product's appearance in the market.
uct is not especially important to con- new-product The real lesson of the failed new Coke
sumers, they will tend to seek out the development process introduction is that consideration of the
opinions of others if they regard the prod- effects of social influence must become a
uct as complex (e.g., personal computers, standard part of the new-product develop-
audio equipment) or if it is a service that is publicity, consumers rated the new Coke ment process. This can best be done by
difficult to evaluate before purchase. flavor labeled as old Coke to be more appreciating that focus groups can reveal
desirable than the old Coke flavor labeled important insights about social interaction
For products deemed high in any of as new Coke, effects that other qualitative methods are
these three factors, the results of group Even if it was later revealed—by Pepsi. likely to miss, and by developing new
interview methods are more likely to be for example—that Coke had secretly research procedures that make possible the
accurate than those of individual interview phased in a reformulation, it is unlikely reliable and representative quantification
methods in cases when the results of the that there would have been any serious of these insights. | ^
two research methods conflict. consumer reaction. Consumers' awareness
For the introduction of new or altered that they had already been drinking the Footnotes
products that are not highly visible or new formula with no decrease in their 'Atihough both the locus groups atid the survey
important and do not involve a difficult enjoyment would probably have made it provided indications that there would be consumer
decision, the marketer can often exert somewhat difficult for them to summon dissatisfaction, the survey results indicated that this
dissatisfaction would be limited to a smalt segment
some control over the degree of the con- the indignation necessary to voice public of the market; the Twus groups suggested the dissat-
sumer's awareness of the opinions of oth- complaints. isfaction would be widespread.
ers. This makes it possible for the mar-
keter to resolve conflicting results of indi- -Case conlributed by Edith Krieger of Psychono-
Conclusions metrics Inc., Southampton. Pa.
vidual and group research by designing
tbe marketing program to create the condi-
tions which produced the more favorable
result.
For example, the drug company that
tested the new birth-control product was
T be Coke reformulation attempt was a
dramatic example bow consumer
awareness of the reactions of other
consumers can play a critical role in the
success or failure of a new or altered prod-
About the Author
Robert M. Schindler is
Associate Professor of Mar-
advised to inform physicians about the uct. It highlights the necessity for some keting a( Rutgers University,
Camden, N.J.
product through its detailers so that each explicit investigation of social-interaction
physician could form an opinion on an effects during concept-testing research.
individual basis, as opposed to launching At the very least, exploratory research
the produet at a iiu-ge conference or other should routinely be carried out using indi-
group setting. vidual as well as group methods. If the
Further, if the marketing program results of the two methods disagree, then

2 6 December 1992
 References
Allvine. Fred C. (1987), Marketing: Principles and Practices. New York: HonomichI, Jack (1985), "Missing Ingredients in "New' Coke's Research."
Harcourt Brace Jovanovich. AdvfiUsing Age (July 22), 1,58.

Alsop, Ron (1985), "Coke's Flip-Flop Underscores Risks of Consumer Taste Kotler, Philip (1988), Marketing Management: Analysis, Planning, imple-
Tests," Wall Street JournaHiuly 18), 23. mentation, and Control, 6th cd. Englewood Cliffs, NJ: Prentice-Hall.

(1987), "New Coke Is a Smash Success With Consumers in This Louis, J. C. and Harvey Z. Yazijian (1980), The Cola Wars. New York: Ever-
Poll," Wall Street Journal {Feb. 26), 23. est House.

Alter, Stewart (1985) "Coke Hurting Research Image," Advertising Age (July Martin, Stephen H. (1987), "Marketing Can ^Make A Good Life Better' If It
Is Properly Practiced," Marketing News, 21 (April 10), 2.
22), 58.

Anderson, W. Thomas Jr. and Linda L. Golden (1984), "Bank Promotion McCarthy. Michael J. (1990), "Coke II Survives One Test City, Heads to Sec-
Strategy," .lournal of Advertising Research, 24 (April/May) 53-65. ond," Wall Street Journal {Oct. 5) BLIO.

Bearden, William O. and Michael J. Etzel (1982), "Reference Group Inllu- __^ (1991), "Coca Cola is Facing New Pepsi Challenge: Avoiding
ence on Product and Brand Purchase Decisions," Journal of Consumer Signs of Age," Wall Street Journal <Ocl. 1), Al,14.
Research, 9 (September), 183-194.
McQuarrie, Edward F. and Shelby H. Mclntyre (1990), "What the Group
Boure, Hrancis S. (1957). "Group Influence in Marketing and Public Rela- Interview Can Contribute to Researeh on Consumer Phenomenology,"
tions," in Some Applications of Behavioral Research, Rensis Likeri and Research in Consumer Behavior, 4, 165-94.
Samuel P. Hayes Jr., eds. Paris: The UNESCO Press.
Oliver, Thomas (1986), The Real Coke. The Real Story. New York: Random
Clark, John B. (1987), Marketing Today: Successes. Failures, and Turn- House.
arounds. Englewood Cliffs, NJ: Prentice-Hall.
Pierce, W. David (1987), "Which Coke Is It? Social Influence in the Market-
Crawford. C. Merle (1987), New Products Management, 2nd ed. Homewood, place," Psychological Reports, 60, 279-86.
IL: lrwin.
Policano, Christopher (1985), -Dueling Colas," Public Relations Journal
Engel, James F., Roger D. Black well, and Paul W. Miniard (1990), Consumer (November), 16-17.
Behavior, 6th ed. Hinsdale, IL: The Dryden Press.
Piiltz, Kim (1985), "Wi/.ards of Marketing," Newsweek (July 22), 42-44.
. . and Robert J. Kegerreis (1969), "How Information is
Used to Adopt an Innovation." Journal of Advertising Research, 9 Quickel, Stephen W. (1989), "Hard Lessons of the Eighties," Business Month
(December), 3-8. (June), 62-68.

Enrico. Roger and Jesse Kornbluth (1986). The Other Guy Blinked. New Ringold, Debra J. (1988), "Consumer Response to Product Withdrawal: The
York: Bantam Books. Reformulation of Coca-Cola." Psychology and Marketing, 5 (Fall), 189-
210.
Feldman, Sidney P. and Merlin C. Spencer (1965), "Ihe Effect of Personal
Influence in the Selection of Consumer Services," in Proceedings of ihe Robertson, Thomas S., Joati Zielinski. and Scott Ward (1984), Consumer
Fall Conference of the Ameriean Marketing Association, Peter D. Ben- Behavior. Glenview, IL: Scolt, Foresman and Company.
nett, ed. Chicago: American Marketing Association.
Schiffmati, Leon G., Joseph F. Dash, and William R. Dillon (1975), "Interper-
Fisher, Anne B. (1985), "Coke's Brand-Loyalty Lesson," Fortune, i 12 (Aug. sonal Communication: An Opinion Leadership/Opinion Seeking Compos-
5). 44-46. ite Approach," in Combined Proceedings, Edward M. Mazze, ed. Chica-
go: American Marketing Association.
Gelb, Betsy D. and Gabriel M. Gelb (1986), "New Coke's Fizzle—Lessons
for the Rest of Us," Sloan Management Review (Fall), 71-76. Scredon, Scott and Marc Frons (1985), "Coke's Man on the Spot,'" Business
Week (My 29). 5(}-6].
Giges, Nancy (1985), "Coke's Switch a Classic," Advertising Age, 56 (July
15), 1,82. Stevenson, Richard W. (1985). "The Revival of the 'Old' Coke," New York
Times (July 12), Section 4, 1,39.
___^__ (1986), "Adman of ihc Year: Coca-Cola's Roberto Goizueta Engi-
neers Startling Comeback," Advertising Age, 64 (Dec. 29), 1,26-27. Tauber, Edward M. (1975), "V^'hy Concept and Product Tests Fail to Predict
New Product Results." ./wwrau/ of Marketing, 39 (October), 69-71.
Greenwald, John (1985). "Coca-Cola's Big Fizzle," Time (July 22), 48-52.
Tongren, Hale N. (1987), Ca.ses in Consumer Behavior. Englewood Cliffs,
Harvard Business School (1978), "Pepsi-Cola (A)," Case # 9-579-108. NJ: Prentice-Hall.

Hendon, Donald W. (1989), Cta.ssic Failures in Product Marketing. New Urban, Glen L. and John R. Hauser (1980), Design and Marketing of New
York: Quorum Books. Products. Englewood Cliffs. NJ: Prentice-Hall.

Hollie, Pamela G. (1985), "Coca-Cola Changes Its Secret Formula, In Use far
99 Years," New York Times (April 24), 1.

December 1992 2 7