How Personality Makes a Difference

Joseph T. Plummer

As consumer researchers we are allprobably pleased about some of the re-cent progress in the area of under-standing and researching consumers. Avariety of approaches and methodolo-gies are now available that examine theconsumer from every conceivableangle. To my mind, however, no ap-proach has been more enigmatic or hasbeen held with such high expectationsas that of personality research bor-rowed from the field of psychology.

The whole concepi of personalityand its relationship to how consumersrespond has always had a tot of intu-itive appeal to consumer researchers.If—as this feeling goes—we can onlyisolate and understand the basic drivingforce that predisposes consumers to actin certain ways, and personality traitsmust be one of those driving forces,then our jobs, of course, could be mademuch easier.

It is understandable, also, that per-sonality researchers in psychology andin consumer behavior have devotedmuch time and attention to developinga whole array of tools: theories thatidentify important personality charac-teristics; scales that classify person-ality traits; and multivariate techniquesthat systematically relate personality

traits to behavior. In some cases, thefervor of these endeavors would evenpale the search for the Holy Grail.

Why, then, has personality researchas applied to people and their behavioras consumers, failed to live up to suchhigh expectations? Why hasn't all ofthis effort had at least some impact onour working models of consumer be-havior and the day-to-day decisions wemake based on these models?

Once we step back away from theparticulars, we begin to get a sense thatpersonality research as it is generallyconceptualized and undertaken hasseveral flaws that litnit its utility.

Conceptually, one of the biggestproblems that has to be overcome ishow to define personality and, impor-tantly, how this concept as defined re-lates to people's lives. One popularworking definition is that personality isthe way individuals react fairly consis-tently to a variety of environmental sit-uations. However, no researcher canever "observe" a person in all relevantsituations to make sure that person isreacting in a consistent manner acrossthe board. Furthermore, we couldquestion whether in fact certain per-sonality traits can be all that stable overtime. It is entirely possible that people

could exhibit some traits when they areyoung—and simply outgrow them asthey mature.

Apart from the problem of definingwhat personality means, the mere pro-liferation of personality batteries andscales points to the fact that we simplydo not have one "best" way of mea-suring it within human beings in a valid,reliable way.

This dilemma presents a real "catch-22" for personality researchers. "Pur-ists" in the field insist on using scalesin which the validity and reliabilityhave been established and docu-mented— and they administer thescales exactly as prescribed. Unfortu-nately, these kinds of scales and bat-teries tend to be developed under"laboratory" conditions—and maynot be the best survey tool for under-standing and segmenting people as itrelates to consumer purchasing andconsumption.

"Pragmatists" in the field believethat the personality measures shouldaddress people as consumers, so theytend to develop their own scales andbatteries. But many of these re-searchers would be hard-pressed todemonstrate that their scales are reallymeasuring the personality traits they

November . December 2 0 0 0 JflUROfIL Of flDUEfiTISIIlG BESEflRCH 7 9

claim the scales are measuring, and—probably due to cost and time con-straints—just do not bother with theproper reliability checks.

If defining and measuring personalityare not insurmountable barriers, thenresearch designs that incorporate per-sonality variables make the utility ofpersonality research questionable.Many studies either implicitly or explic-itly assume that somehow certain per-sonality traits can lead people to be-have in certain ways.

Many empirical studies have beencompleted based on this premise. Yetin the final analysis—when al! of themethodological and statistical geegawsare stripped away—we have, at best,a "direct l ine" relationship thatamounts to about 10 percent of thevariance in behavior. Obviously, theother 90 percent reflects the vast com-plexity of people, and the sum total oftheir being, of which personality is onlyone part.

Major efforts over the years to pro-file buyers versus non-buyers, such asthe classic work by Evans in the early1960s where he attempted to profileChevrolet owners versus Ford ownerson numerous personality traits, or therecent personality scales in the Sim-mons media and market surveys, havemet with mixed results, at best, pri-marily for two reasons: First, theywere never based upon any strongtheoretical consumer model — theywere merely research profiling exer-cises across accepted personalitytraits. Secondly, most researchers andalmost all marketing or advertising de-cision-makers were unsure how to in-terpret or translate the results intoaction.

The most vivid example of which Iam aware in the area of personality seg-mentation research was conducted byProcter and Gamble In the early 1970s.As you would expect, P&G ap-proached the research with care, dili-gence, and some of the best researchminds in and outside the company theycould utilize. After three years of ef-fort, it was finally abandoned becausethe brand and advertising managers

could not effectively deal with results.A valiant but unsuccessful effort.

But do not lose hope—there is anarea of personality research which we(and I am sure others) are working withthat is useful and can make a differ-ence. It is the personality of brands—not individuals. At Young & Rubicamwe have been examining the issue ofbrand personality for a number ofyears. There are in reality two differentfaces of brand personality. It Is neces-sary to understand both faces in orderto better grasp the totality and powerof this useful strategic concept.

We found initially that there was

Joseph T. Plummer is executive vicepresident and director of research ser-vices at Young & Rubicam USA, and isa member of the board of directors. Heis based at Young & Rubicam NewYork.

Prior to becoming Y&R's research di-rector in March 1979, Mr, Plummer hadbeen with the Leo Burnett Co. in Chi-cago where he was senior vice presidentand research director.

Dr. Plummer holds a B.A. degreefrom Westminster College (PA) andhoth M.A. and Ph.D, degrees in com-munications from Ohio State University.He is on the editorial boards of theJournal of Marketing, Journal of Adver-tising, Journal of Advertising Research,Journal of Broadcasting, and Journal ofBusiness Research. He is a frequent con-tributor to these and other research andmarketing publications.

some semantic confusion and differ-ences in terminology to be cleared up,not least of which Is the difference be-tween brand image and brand person-ality. Let's talk about brand image first,A brand presents itself to the worldin many ways—through the productitself, through its packaging, itsname, where it is sold, A brand sold ina supermarket is communicatingsomething.

The world, on the other hand, inter-prets the brand through many differentfilters; through experience, throughperceptions, misconceptions, the valuesystems of the individuals out there inthe world, and, of course, all the noisein the system. Because if there is onething we know, it is that transmissionsare rarely clear and unambiguous.

The two faces of brand personalitytherefore are input, that is, what wewant consumers to think and feel, andout-take, what consumers actually dothink and feel. These two perspectiveson brand personality can be expressedin two forms. The first is the brand per-sonality statement, that is. our com-munication goals for the brand whichhave been in use at Y&R for manyyears as an important part of creativestrategy. And the other is the brandpersonality profiles, which are con-sumer perceptions of the brand.

As we all know, any brand can bedescribed in terms of three differentclasses of characteristics. The firstconsists of its physical attributes. Tangis an orange powder that comes in ajarand costs 69 cents. These physical at-tributes of the brand are independentlyverifiable. That is, an individual cansee for him or herself exactly what thecharacteristics of the brand may be.The second way of describing a brandis in terms of its functional character-istics or the consequences of using abrand. Lemon-freshened Pledge pol-ishes your furniture. It repels dust.These kinds of consequences of branduse are also objectively verifiable.However, if you use lemon-freshenedPledge, it may also have the conse-quence of communicating to you or toothers that you are a good homemaker

8 0 JOURflHL OF flDUEIITISIIlG flESEflflCH November . December 2 0 0 0

or a careful shopper. Thus, the func-tional consequences of using a brandare in some cases external functionsand in other cases have internal effectson the head of the person who is usingthe product. The third way of de-scribing a brand is characterizational.That is, a brand may be characterizedas modern or old-fashioned, or lively,or exotic. These characterizational as-pects of the brand are what we call thebrand's personality. The characteriza-tional aspects of the brand or its per-sonality are purely the result of com-munications because there is rarelyanything intrinsic to a brand that makesit lively, or exotic, or sophisticated,

There are a lot of different ways ofdescribing how advertising works, andthere are all manners of complexmodels that various people have ad-vanced to objectify this process. Myrather sitnplistic view is that the con-sumer who favors or uses any partic-tjlar brand most often has looked at allthis information and communicationthat we have, as advertisers and mar-keters, put out about the brand. Andsomewhere inside his or her head he orshe has said, I see myself in that brand,or I see that brand in myself. A non-favorer, on the other hand, has lookedat all this information that is out thereabout the brand and concluded thatthat brand is not for him. He does notsee himself in that brand or that brandin himself and, consequently, he is anon-favorer of that brand and a favorerof some other brand. Brand personalitycan play a key role in the "for me"choice,

So then, to recapitulate, there arethree primary components to a brand'simage, three aspects of the brand's de-scription. There are its physical ele-ments or attributes, the functionalcharacteristics or the benefits or con-sequences of using a brand, and theway the brand is characterized, or itspersonality. These elements, mediatedby whatever the viewer brings to theinteraction, are in some way trans-formed into the viewer's head into "ap-propriate for me" or "not appropriatefor me"; or possibly, "me for it."

The brand personality statement isprimarily a strategic tool for the cre-ative. It is written by the creatives andis used by them. It is used to sotne de-gree as a mirror to hold up to theirwork, to sec whether the ads or com-mercials that are produced are conso-nant with the personality as expressedin the brand personality statement. Itis an articulation of what we would likeconsumers out there in the world to feelabout our brand over time.

The brand personality consumer pro-files are quite different. Their role andfunction is different. The way they arewritten is different, and the way theyare measured and evaluated is dif-ferent. While brand personality state-ments have been in use at Y&R forover 10 years, the profiles are a recentresearch development.

The goal of the brand personalityprofiles is to describe perceptual realityfrom the consumer perception. That is,they should reflect the way consumersactually feel about the brand ratherthan simply being an expression of theway we would like consumers to feelabout the brand. The profiles shouldidentify salient components of thebrand's personality. That is, theyshould indicate which dimensions arestrong and which dimensions are weak.The profiles should be used to comparepopulation subgroups because if thereis one thing we know about any numberthat we get through research, it is thatthat number alone is wrong and oftenmisleading. If we do a piece of researchand get something that says 35 percent.we know that in reality it isn't 35 per-cent. It is more or less 35 percent oraround 35 percent, but not precisely 35percent, in order to understand whatthat 35 percent means, we have to com-pare it to something else. If 35 percentof the population says that a brand ismodern, this has a different meaning ifit is examined in the context of a 55percent for old-fashioned than in thecontext of a 15 percent for old-fashioned,

Brand personality profiles tend to beportrayed or written in graphic formbecause it is easier to understand re-

lationships when you see charts ratherthan an array of numbers. But still,you've got to have some numbers toback up the charts so you know whatthe charts refer to, and. finally, as in-dicated, they should be written In termsof comparisons; comparisons betweenrelevant subgroups, or cotnpaiisons ofthe profiles of one brand to those ofanother.

The brand personality profiles aremeasured through consutner surveys inmost cases. We have developed somerelatively simple checklist proceduresto give the actual measurement,

I'd like to briefly share with you theR&D we went through to make the pro-files an operational tool at Young &Rubicam. We did some early develop-mental work on brand personality, in-tending to see whether simple checklistprocedures could discriminate betweenbrands. We started off with a relativelyeasy task. That Is to say, if the proce-dure could not differentiate betweensome well-known brands like MillerHigh Life. Holiday Inn, and Oil ofOlay, we'd have to go back to thedrawing board and find some other pro-cedure. So we developed a 50-attributechecklist from previous research onpersonality and in-depth interviews.We asked respondents to indicatewhich of the words and phrases on ourlist they would use to describe each ofthe brands we were interested in. Whenwe looked at the data it was clear thatthis procedure did indeed discriminate.

Taking a look at the first attributewhich was "cheerful," we found that39 percent of the sample said HolidayInn was "cheerful," while only 6 per-cent said Birds Eye was "cheerful."Forty-two percent described Atari as"youthful" and only 3 percent de-scribed Holiday Inn as "youthful."Thirty-nine percent described Oil ofOlay as "gentle" and 0 percent de-scribed Miller High Life as "gentle."Thus it was clear that this checklist diddiscritninate by looking at the data hor-izontally from brand to brand. Lookingat the profiles of the different brands,they also seem to make sense. HolidayInn was described as "cheerful."

November • December 2 0 0 0 JOURdflL OF flDUEIlTISlOG RESEflRCH 8 1

"friendly," "ordinary," "practical""modern," "reliable," and "honest."Oil of Olay was described as "gentle.""sophisticated," "mature." "exotic,""mysterious." and "down to earth."So the procedure that we were using atleast met the relatively simple criterionof being able to differentiate betweensome major brands across differentproduct categories.

Since brand personality is largelysymbolic, we thought it would be in-teresting to play around with somesymbols. So we asked respondents toplay a game with us. You have allplayed the game which goes: "If youwere a vegetable what kind of vege-table would you be? If you were an an-imal, what kind of an animal would yoube?" We asked the respondents to de-scribe the brands in these terms. Wegave them a list of 29 animals and said:"If each of these brands was an animal,what one animal would it be?" We alsoshowed them 25 different activities, 17fabrics, 35 occupations, 20 nationali-ties, and 21 magazines. When we hadfinished this process, each respondenthad created his or her own uniquesymbol for each of the brands. Thatsymbol was a particular animal, a par-ticular activity, a fabric, an occupation,a nationality, and a magazine. We thenasked the respondents to describe thiscomplex symbol that they had createdby means of the same list of 50 attri-butes that we had used to describe thebrands. Indeed, respondents werequite able to do this test.

When we take a look at the averagesymbols that were created, that is, themost frequently mentioned animals,countries, occupations, and so forth, arather interesting picture emerged. ForOil of Olay, for instance, the animalwas mink, the country was France, theoccupation was secretary, the fabricwas silk, the activity was swimming,and the magazine was Vogue. It doessort of bring to mind the picture ofsomeone's secretary on the Riviera, bythe swimming pool, in a silk bathingsuit, reading Vogue, with her mink coaton the adjacent chair. This was quite

different from the symbol for KentuckyFried Chicken, which was a PuertoRican zebra, a housewife dressed indenim, camping, and reading TVGuide. When we looked at those data wesaid, "Zebra, why should KentuckyFried Chicken be a zebra?" We didn'tknow. So we went to the product groupand said, "Why is Kentucky FriedChicken a zebra?" And they said,"Look at the package, look at thestripes." So it seemed apparent thatthere was indeed some reasoning be-hind the characteristics that the re-spondents were ascribing to the brand,at least in these symbolic terms.

I said earlier that our simple modelof the advertising process is one whichsays that consumers see all the infor-mation about the brand that we are put-ting out and either do or do not seethemselves reflected in that brand.When we take a look at who consumersthought the stereotypical user of Oil ofOlay was, that is, how they would de-scribe a woman who uses Oil of Olayrather than some alternative facialmoisturizer, that woman was describedas "mature ," "down-to-earth,""calm," "honest," "practical," andso forth. In short, the stereotypicaluser of Oil of Olay is a pretty down-to-earth, solid female citizen. Whereas thepersonality of the brand was somewhatmore upscaled and aspirational. This isprobably quite appropriate for a cos-metic since, at least to some degree,women buy cosmetics to reflect whatthey would like to be rather than whatthey actually are.

I would like to share a recent casehistory where brand personality andthe brand personality profiles played amajor role in advertising development.The brand is Dr Pepper. But to fullyappreciate the case, we need to take astep back in time.

During the 1960s Dr Pepper's rate ofgrowth had been better than doublethat of the industry and of its majorcompetitors. In 1969, it was in fifthplace nationally in case sales behindCoke, Pepsi, 7-Up. and Royal Crown.Its growth had stemmed obviously

from increased availability nationally,but of equal importance was its con-tinued share growth in the Southwest.

Consumer research undertaken atthis time indicated that there was a highdegree of awareness (on an aided basis)that Dr Pepper was Indeed a soft drink.However, there were the following mis-conceptions about the brand: It is madefrom prune juice: it contains peppers orpepper sauce; it is medicinal; it aidsregularity; etc. Even in the Southwest,where the brand was second only toCoke in sales, some of these myths per-sisted. These misconceptions mademany people reluctant to try DrPepper. Also, first-time triers wereoften disappointed because they hadnot been properly prepared for its tastedifference (because of its brown color,many expected it to taste like a cola).Lastly, Dr Pepper was heavily outspentin media by Coke, Pepsi, and 7-Up. Asa result brand perception, power, andrecognition were low. Based upon anal-ysis of this research, along with othermarket data, the advertising objectivesthat followed seemed obvious: con-vince young experimental prospectsthat Dr Pepper is a deliclously dif-ferent-tasting soft drink. However, the"how" was more difficult.

The campaign that emerged has sincebecome a part of advertising history.The slogan that was created was: "DrPepper—America's Most Misunder-stood Soft Drink." The copy initiallymet the misconceptions head-on andused them as a handle in explaining thatDr Pepper is a good-tasting, unique softdrink. Originally, "misunderstood"commercials positioned Dr Pepper asan underdog fighting to gain awarenessand used company spokesmen settingout to dispel the misconception andgain new "triers"—in fun, and attimes, irreverant and larger-than-lifesituations. Over the years, the cam-paign evolved so that the companyspokesmen were replaced with recentconverts attempting to convince "hold-outs" to try Dr Pepper. It was suc-cessful not only in building a bright newcharacterization for Dr Pepper, but also

8 2 JOURnRL QF BDUERTISinG BESEflRCH November . December 2 0 0 0

in terms of adding to sates increases,Dr Pepper, from 1%9 to 1974, grew ata rate better than double that of the in-dustry. In 1973 it passed Royal Crownin sales, moving into fourth position na-tionally. By 1974, top-of-mind aware-ness had increased dramatically. Thebrand was in fourth position in salesand moving up on third place.

In truth, Dr Pepper was no longermisunderstood. It had come into itsown. As successful as the '"misunder-stood" campaign had been, it was timefor a change. With the change inmarket position, the misunderstoodstance was no longer appropriate; itwas time for Dr Pepper to behave andtalk like the contender that it had be-come. It seemed only logical tbat thenew advertising should be more posi-tive and even more aggressive.

However, many of the factors thatoriginally made Dr Pepper misunder-stood were what had made Dr Pepperunique. No matter how understood Ithad become, Dr Pepper still had themost distinctive taste of any soft drink,the most distinctive name of any softdrink, and the most individual person-ality of any soft drink. In the soft-drinkspectrum, it stands alone as a true orig-inal. In the words of the Y&R creativesupervisor at that time Dr Pepperwas—"The Most Original Soft DrinkEver," And that, of course, is the waythat the new advertising positioned thisunique drink: as "The Most OriginalSoft Drink Ever." Naturally enough itwas hoped that "The Most OriginalSoft Drink Ever" should have the mostoriginal advertising ever.

An effective medium for telling sto-ries with music is the Broadway mus-ical. Accordingly, the agency creativeteam created five mini-Broadway pro-ductions for the television campaign,and productions are what they were.Even though they were all subse-quently filmed on a theater stage, theywere all different. Each in its own rightwas a story about Dr Pepper.

Dr Pepper continued to see con-stimer response and growth. Thelaunch of Sugar Free Dr Pepper, to cap-

italize on the rejuvenation of the dietsegment in the mid-seventies, also pro-vided additional growth. With its newlyfound status of a major factor in thesoft-drink industry, Dr Pepper began toclose in on the third-ranked brand 7-Up, and a strategic change was madein 1979. The strategic shift was de-signed to capitalize on Dr Pepper's per-ceived mainstream position and tofocus on its popularity. After all, thelong-term goal was to double its size by1985, the company's 100th year. Theagency created a new campaign fea-turing David Naughton meeting "Pep-pers" everywhere he danced across theUSA, The "Be A Pepper"—originallike a Pepper was launched. But thenDavid decided to become a werewolfin London and new management fromthe world of package goods arrived inDallas. The new management re-quested a retrial strategy for Dr Pepperand a completely separate positioningand campaign for Sugar Free DrPepper.

In the midst of the Coke-Pepsi war,the launch of Sunkist, Mountain Dew,and a slew of other brands targetedagainst the Dr Pepper youth franchise,growth for Dr Pepper had come to ascreeching halt! Several of us believedthat beyond what was going on com-petitively, we had Inadvertently walkedaway from Dr Pepper's majorstrength—its unique brand personality.That was what really set it apart forcertain consumers as an alternative toCoke and Pepsi and fruit flavors. Theproblem was how to convince Dr Pep-per's management of this matter. Whatwe decided to do was do small-scalebrand-personality research on our own.

In the Dr Pepper research we con-centrated on the brand-personality gapbetween favorers and nonfavorers toadd support to our original hypothesisabout the "Dr Pepper problem." Wecould see from the results that the per-sonality of the brand that was in theheads of these two drinker populationswas quite different. There was reallyvery little relationship between thesetwo images. The question that we

asked ourselves was: Is this simply acharacteristic of the product category,that there is no real similarity in theway these two different categories ofconsumers perceived the brand?

When we look at the image of Coke,as held by favorers and nonfavorers,we see that by and large they matchvery closely. There is a difference inlevel, but for most items the brand isperceived very much the same by fa-vorers and nonfavorers.

We also examined the image of thestereotypical user of the brand, andonce again we found that people's per-ception of the kind of person who usesDr Pepper is quite different when youlook at tbe perceptions of favorers andnonfavorers of Dr Pepper, For Coke,once again, the image is exactly thesame. Both favorers and nonfavorershave the same perception of what arethe characteristics of the typical"drinker,"

We concluded from this, and theclient concurred, that our loyal usersstill retained the original, fun, off-beat,and underdog personality the adver-tising had built up over the years. Butother consumers exposed more re-cently to the newer advertising withchanging campaigns, multiple mes-sages, and a more mainstream person-ality perceived Dr Pepper differently.We had to get back to the real brandpersonality in our advertising, but ex-press it in a radically fresh way.

In summary, we have the brand-per-sonality profiles, the brand-personalitystatements, and the procedure that weuse to integrate the two in order toknow what we want to maintain, whatwe want to change, and how best to goabout doing it. Therefore, while thesuccess of personality segmentation onconsumer populations to create per-sonality typologies or unique user pro-files using survey methodology has notproven very useful to marketers andadvertising people, we, at any rate, arevery encouraged by our work withbrand personality. It is in this newarena of brand personality that re-search does make a difference!

November . December 2 0 0 0 JQUflnRL OF flDUEflTISinG flESEBRCH 8 3