Welcome to this inaugural issue of the ConsumerPsychologist Newsletter!  Every month, I will feature a concise discussion of current events and issues in consumer marketing.  Newsletters will contain words of wisdom and marketing insight, but will never exceed three pages in length—thus there is no need to put off reading it!

CURRENT ADVERTISEMENTS USE “TWO-SIDED” APPEALS.  Psychologists have known for a long time that either (1) admitting something negative about your own product or (2) admitting something positive about a competing product may significantly improve your credibility.  This technique tends to be especially effective when this admission is followed up by positive information about your product.  For example, Professors Michael Kamins and Larry Marks ran an experiment in which they exposed subjects to two different kinds of supermarket comparison ads.  One was of the traditional kind, showing the sponsoring supermarket to have the cheapest “basket” among all the supermarkets compared.  In another condition, however, one competing supermarket was actually shown to have lower prices.  The ad then went on to say, however, that the lower prices in the competing market were by far outweighed by greater service in the sponsoring supermarket.  Interestingly, the supermarket was rated more favorably when it admitted that a competitor had slightly lower prices.

Marketers seem to be reluctant to use the two-sided appeal strategy since this can be risky.  In some cases, the strategy backfires, and the consumer forms a more negative opinion of the product.  This appears to be especially the case for low involvement products, where the consumer may merely notice that something negative is being said while not taking the time to explore the importance of this problem and “compensating” factors. 

One classic advertisement that was run back in the early 1980s challenged the reader to be aware of the disadvantages associated with a portable phone that was advertised.  The ad admitted outright that the user could “Forget about dialing!”  The ad then claimed that adding a dialing feature would have increased the price by several hundred dollars but that this much cheaper unit could be used to answer calls.  This allowed the user to avoid missing any calls, which was the most important benefit of the product touted.

Recently, some advertisers have given the two-sided appeal a try.  One of my favorite ads is the current one for Listerine.  “You can do it!” the ad starts out.  It acknowledges that “Sure, it’s a little intense.”  The conclusion is that “You can handle it, but the germs can’t!”

In these days where many ads try to draw attention away from the price of a product, one ad for a diet supplement asks incredulously, “Is any bottle of pills really worth $153?”  The ad then goes on to say that this supplement is “much too expensive”—and “much too powerful”—to be used by those who are only slightly overweight.  It should be used, the ad says, only by those who are seriously overweight.

Subscribe to the Consumer Psychologist Newsletter!

Do these two-sided appeals work?  It is difficult to predict how well any ad will be received—and especially so for two-sided appeals—before the ad is “tested” on consumers who give feedback on their perceptions.  The makers of Listerine, with the large budget necessary to run all these ads, have almost certainly tested the ad and found it to be effective.  Interestingly, the ad seemed to be shown sporadically last summer, and has appeared much more frequently since December.  Results may have been impressive!

DO SALES SEEM TO DISAPPEAR SOONER THESE DAYS?  There have always been stores that have offered periodic sales that last only one day or a few days.  Such sales, however, seem to have been on the rise recently.  There are a variety of reasons why this may be happening.  One big factor is the current economy.  Many retailers feared that 2002 Christmas would be disappointing—and most were justified in this fear.  This is probably why a number of sales seemed to “creep” up early—as early as the Veterans’ Day week-end.  Such sales, however, introduce a problem in that they may provoke competitors to retaliate with sales of their own.  This can be a vicious circle, where retailers begin their sales earlier and earlier—and with steeper and steeper discounts—to avoid being outdone by competitors.  (Those familiar with game theory will recognize this as an example of the “Prisoners’ Dilemma”).  A brief sale, however, seems less provocative, and demonstrates to competitors that the intent is not to start a price war.  Short term sales also allow for price discrimination:  Price sensitive shoppers will tend to go out to take advantage of the sale even if this is inconvenient.  More convenience oriented consumers, in contrast, will pay higher prices to shop when they want to.  Stores have a great deal of fixed costs, so it is often worthwhile, once the fixed expenses have been “paid for” by full price shoppers, to settle for a smaller profit on customers who would not have bought at full price.   K-Mart seems to be running a lot of two-day sales at the moment.

DO CUSTOMERS REALLY KNOW WHAT THEY WANT?  In one song, Diana Ross asks the probing question, “What are you hoping for?  Do you know?”  Consumers are not stupid, but understanding what is really important to us is often difficult.  Sometimes, we may recognize certain features to be very important to us, but we may have difficulty deciding which features are most important and which can be “sacrificed” for other, more important features.

Consider, for example, the case of a consumer shopping for a car.  Chances are, when you ask how important safety is, the consumer will say “Very important!”  How important is reliability?”  “Very important!”  Gas mileage?  “Very important?”  Performance? “Very important!”

Even so, you might think to ask consumers to rate the importance of each attribute on a scale of 1-10, with 10 being most important.  Here, consumers may actually rate, say, safety as a 10 and gas mileage as 9.  However, the difference here is modest, and such responses are frequently not consistent.  The consumer might reverse the two ratings the next day.

Another option involves asking the consumer to “distribute” one hundred points between the features according to their importance.  The consumer, then, is asked to trade off attributes against each other.  If one is rated very high, others must be “sacrificed.”  For example, the consumer might rate gas mileage as 20 and safety as 40.  Most consumers, however, find this task very difficult, and results from such “fixed sum” scales tend not  to be reliable.

Is there a better way?  Yes, but the exact method may surprise you.  It is quite counter-intuitive.

Conjoint analysis asks the consumer to rate his or her overall valuation of a series of complete products—cars in this case.  The respondent may be shown a “profile” of a car that has “Above average” reliability, gets 20 MPG, has an “Average” safety, and costs $17,000.  The consumer then goes on to rate other combinations, such as “Below average” reliability, 25 MPG, $21,000, and “Average” safety. 

A statistical algorithm is now used to “decompose” the consumer’s evaluations into importance scores for each attribute.  Consumers may then reveal, for example, that reliability is twice as important as gas mileage, for example.  Several conjoint methods—and software packages—exist which differ somewhat in their criteria and methods.  Most use techniques similar to multiple regression and analysis of variance.

Different groups of consumers often differ in the weight given to each attribute and its preferred state.  That is why some people shop in Wal-Mart and others shop at Nordstrom’s.  Conjoint analysis can identify different segments and the attribute importance and preferences of each.

Conjoint analysis does have several drawbacks.  Generally, very large sample sizes are needed to get precise results.  While the required sample size depends on the number of attributes and the number of “levels” each can have (e.g., 20, 25, and 30 MPG would represent three levels of gas mileage), a minimum sample size often runs in the hundreds.  Having multiple segments whose preferences have to be untangled can also “explode” the sample size needed.   It may also be necessary—if a lot of attributes and levels are involved—to have each subject fill out numerous profiles.  This can be exhausting, and may cause either incomplete questionnaires or low quality answers.