Printer Friendly

Marketing Benchmarks: Do You Trust Your Friendly Marketer?

How much toothpaste should one put upon the brush? How often should one change the car's oil? How much laundry detergent is the right amount? How much should someone spend on an engagement ring? Should the kid get braces? How frequently should one get a medical checkup? Consumers face questions such as these daily. How they determine answers depends on a number of factors. How important, for example, are the associated costs of a poor decision? Changing the car's oil every five hundred miles might be better for the car's engine than changing it every five thousand miles, but the benefits might not justify the additional time and money. A ten thousand dollar engagement ring may attract more wide-eyed compliments than a one thousand dollar ring, but only by ignoring other needs of a young married couple. Closely related to, and perhaps a result of the costs, is the effort consumers are willing to put into answering the question. A person might conduct an experiment with laundry over time, using different amounts of detergent, controlling for the types of clothes, to reach a conclusion about the optimum dose, but typically people would not be willing to invest such an effort. Another consideration is how much expertise consumers have. Dentists, perhaps, know that the role of toothpaste is largely to provide fresh breath. They know that too much use of the abrasives in the paste cleans the teeth only in the sense of grinding away healthy enamel. Most people, however, understand little about tooth structure and believe toothpaste is necessary for decay prevention and that more is probably better than less. Automotive engineers may have special knowledge about oil change frequency, but most of us have no clue.

So, most consumers are neither sufficiently involved nor sufficiently expert with the products to invest in research to determine optimal use amounts, optimal expenditures, or optimal replacement schedules. As a result, they likely make these decisions in various ways, including trial and error, satisficing, heuristics, and accepting benchmarks from others. Reasonableness or logic may not matter when consumers make such judgments. Since Tversky and Kahneman (1974), we have recognized that people are subject to adjustment and anchoring biases even from logically irrelevant information. Marketers, of course, are happy to provide suggestions--in advertising pictures (the inch or longer squeeze of toothpaste on the brush), in packaging text ("For best results, use 1/2 cup per load") and in personal selling ("The usual expenditure for an engagement ring is two month's salary"). Herein, I use the term marketing benchmark for a marketer-supplied indication of how much of a product to use, how often to replace it, or how much to spend on it.

Marketing benchmarks are suggestions provided when there is no objective information available to help consumers decide from a range of possibilities. Marketers provide a sticker price, but the sticker price is not a benchmark because it is not a suggestion; it is fixed. The salesperson's suggestion that $50-$75 is the appropriate amount to spend on a wedding gift is a marketing benchmark. Marketers instruct to put two AA batteries in a flashlight, but that instruction is not a benchmark because there is no latitude for variation. The advice to use four ounces of shampoo per use is a benchmark. And the phrase "Time to buy more XXX nuts!" at the bottom of a can of salted nuts is not a marketing benchmark; it is tied to the clear exhaustion of the current supply. But the indication to buy new shocks because the old shocks have fifty thousand miles on them is. Marketing benchmarks are provided to influence the answers to questions when the answers are not clear: How much of a product should I use per application? How frequently should I replace a product when product performance is not a clear indicator? And how much should I spend on a product in a category with a broad range of prices?

Should consumers trust these benchmarks? Two business principles encourage a belief in the accuracy of benchmarks. These principles are well established and may appear self-evident. One is the presumed marketing goal of consumer satisfaction. Kotler (1967) identified consumer satisfaction as the distinctive characteristic of a new business paradigm, marketing, dating back to the work of Theodore Drucker in the 1940s and '50s. Drucker is oft-quoted: "The aim of marketing is to know the customer so well that the product... sells itself (Drucker 1973).

If satisfaction is the goal, marketers should advise consumers to use just enough products, just long enough, and at just the right price--all for optimal value. Critics, of course, would argue that there are wide ranges of product use, replacement frequency, and expenditure level that are satisfactory. And even if advertised benchmarks are within the acceptable range for satisfaction, they are apt to be on the high side, with the result that consumers may be using, for example, up to five times as much toothpaste as appropriate or changing oil twice as often as necessary.

A second reason that might be proposed for trusting marketing benchmarks is competitive pressure. The very notion of competition is that firms seek any source of advantage. If a firm can improve quality or charge less, it will gain appeal in the marketplace. (For discussions of various competitive strategies, see, e.g., Kotler and Keller 2016 or Porter 1980). A firm that promotes using too much product, for instance, should be leaving room for a competitor to promote itself as more efficient. Consider motor oil, one of the classic products for marketing benchmarks. The advent of synthetic alternatives has provided an alternative competitive position--more expensive but longer-lasting ("As conventional oils break down, their ability to prevent engine wear diminishes. Mobil 1 motor oils, on the other hand, retain their wear protection properties for a much longer time, increasing engine life" from https://mobiloil.com/en/article/car-maintenance/learn-about-motor-oil-facts/synthetic-oil-vs-conventional-oil). To the extent that few significant brand differences exist, competitive pressure should encourage honesty in marketing benchmarks. Firms that set benchmarks very much in their own favor should be making themselves vulnerable.

There are, however, likely barriers to the effectiveness of competitive pressure. One is brand associations; many products that may have few functional distinctions have other brand elements that may protect them against the risk of excessive benchmarking. Consider Arm & Hammer's Fridge-N-Freezer--baking soda for removal of odors from refrigerators and freezers. The package recommends replacing the box every three months or sooner. It is unlikely that Fridge-N-Freezer is vulnerable to another brand that promotes that it lasts "a full six months." Arm & Hammer's brand is simply too strong and has too much name awareness and trust established.

Another limit to competitive pressure is that a more consumer-friendly benchmark may provide indefensible advantage. Competing on the basis of benchmarks is similar to competing with price discounts. As long as cost structures allow, competitors simply match price cuts. Thus, benchmark-competition is similar to a Prisoners' Dilemma problem--urging consumers to spend less or use less or buy less frequently would clearly save consumers money, but if all competitors did so, it would reduce profit for the entire industry. Few firms are willing to work for an indefensible advantage.

Other factors that might make benchmarks more trustworthy are watchdog individuals or groups and objective opinion leaders. With the growing importance of social media pressure, one might expect overly marketer-friendly benchmarks would be undermined through publicity. An Internet search on any of the examples in this paper will suffice to show that the web is replete with criticisms; yet, the benchmarks persist. Perhaps the power of brands exceeds the power of independent critics. Overall, there appear to be business principles in favor of trusting marketing benchmarks but also compelling arguments against any complete prescription from those principles.

The efficient information market hypothesis (Darby and Kami 1973; Ford, Smith, and Swasy 1988, 1990; Nelson 1970) implies that consumers should be reluctant to trust benchmarks, just as they should be reluctant to believe advertising claims in general. Nonetheless, research on skepticism toward advertising has consistently reported that consumers do use advertising even though they simultaneously report that they do not believe advertising claims (Calfee and Ringold 1994; Obermiller and Spangenberg 1998; Obermiller, Spangenberg, and MacLachlan 2005). It seems likely, then, that consumers may use benchmarks, too, regardless of their objective truthfulness or consumers' own opinions of them.

In support of the notion that marketing benchmarks are untrustworthy, consider a few informal criticisms: A recent Wall Street Journal article on the question of laundry detergent benchmark (Ziobro and Ng 2013) stated, "For years, consumer-product makers could count on extra sales from shoppers who poured in too much detergent with every load." Adam Lowry, a detergent executive, observed, "In an extremely mature market like laundry, for established players to grow they have to either steal share or get people to use more. They are trying to dupe people into using more product than they need." Another appliance expert, Vernon Schmidt, says the proper amount of detergent is as little as 1/8 the recommended amount. Click and Clack of NPR's Car Talk have addressed the question of the frequency of auto oil changes on several shows, noting that most manufacturer's manuals advise oil changes between five thousand and ten thousand miles, not the three thousand recommended by oil change companies and motor oil brands. Numerous websites report that insurance companies have little incentive not to encourage people to purchase more insurance than an objective assessment would suggest (see, for example, Larkford [1990]).

Do consumers use marketing benchmarks? Given the common use of benchmarks in advertising, we can probably safely presume that practitioners believe the answer to be yes. Although academics have not addressed marketing benchmarks as defined here, research has demonstrated anchoring effects of marketer information that is similar to benchmarks. Wansink, Kent, and Hoch (1998) looked at the effect of the number of packages in a sales promotion on purchase decision. The phenomenon was not a direct benchmark of how much was appropriate to buy, but something similar. They found that, when promoting multiple package purchases ("n for $... " pricing), specifying higher quantity limits (e.g., "Limit of 4 per customer") or suggesting a higher purchase amount ("Snickers bars--buy 18 for your freezer"), with all three tactics, higher numbers increased the amount of product purchased or intended to purchase. They also found, however, that this effect was eliminated by having people think about how many products they usually buy at one time. Wansink (1996) separately demonstrated that the size of the package influenced the amount people used--larger packages led to more product use--when product use was discretionary and unconstrained (e.g., for detergent but not for bleach). The Wansink studies are relevant because they show that marketers can subtly affect buyer behavior by suggesting how much people should buy or use, but they do not address the effect of benchmarks on the considered decisions consumers make about how much to use, how much to spend, and how often to replace.

How might marketing benchmarks influence consumer decisions: Do they provide useful information? Is using benchmarks rational? As defined, marketing benchmarks are provided when no objective information is available. At least, not reasonably available. Traditional economics would suggest that benchmark information be evaluated on a cost/benefits basis only, and rarely used by skeptical consumers who see high costs in confirming their validity. Yet, as noted above, countless studies since Tversky and Kahneman (1974) have demonstrated that even random numbers can influence decisions in such contexts. From a behavioral economics perspective, it is understandable that marketing benchmarks are used. Given bounded rationality, high search costs, and often little motivation, it may be entirely rational for consumers to use benchmarks, even if they suspect them to be marketing manipulations.

Consumers may also be engaged in rational satisficing. Faced with a complex decision for which an optimal solution is not worth the costs of information search, consumers may simplify it to acceptable and unacceptable outcomes. If consumers judge the benchmark likely to lead to an acceptable outcome, they could be regarded as rational in using it.

In a model of external consumer search, Schmidt and Spreng (1996) identified clusters of antecedent variables that affect the likelihood of search by influencing perceived ability to search, and perceived costs and benefits of search. They propose that benefits and costs influence motivation to search, which is further moderated by consumer characteristics: enduring involvement, need for cognition, and shopping enthusiasm. A modification of their model might include use of marketing benchmarks as an alternative to search. Ability to search and benefits of search would reduce likelihood of using benchmarks, costs of search would increase use of benchmarks, and additional consumer characteristics might include trust in marketing and trust in the brand.

To recognize the rationality of using marketing benchmarks is not, however, to indemnify them from the perspective of consumer welfare. Consumers are likely overspending by making what they think are good decisions. An example is the use of high-octane gasoline by consumers who think it will enhance or maintain engine performance in their cars. Most cars gain no objective benefit from the more expensive fuel. Many such consumers are likely uncertain if the higher octane actually helps, but they choose to pay the additional 5%-10%, willingly and hopefully, believing that it may help, and further believing that solid information about its benefits is too difficult to obtain. Rational behavior within the context of the decision is not the same as and does not necessarily lead to optimal behavior within the larger context of life.

Research Questions

Two related questions are addressed: Do consumers use marketing benchmarks? If so, in which direction are they biased--toward (even beyond) or away from the marketing benchmarks? And, second, does general advertising skepticism influence individual response to marketing benchmarks?

Preliminary Investigation

The research began with three focus groups designed to identify consumer awareness of benchmarks and confirm the sense of the essential research questions. Subjects (total n = 31) were MBA students and university staff, ranging in age from 28 to 57. The purpose was to investigate awareness of and reactions to marketing benchmarks. The sessions began with an introduction of the concept and display of two ads with benchmarks, one for Colgate toothpaste, picturing a portion of toothpaste just longer than the bristle part of the brush, and a second, for Monroe auto shocks, that explicitly advised to replace shocks after fifty thousand miles.

All participants expressed awareness of benchmarks, and many provided other instances. Some of the examples were arguably simple quality claims, not benchmarks, but, all three groups agreed that marketing benchmarks are a common advertising tool. Initial opinions were split about their purpose--ensuring satisfaction vs. attempting to sell more products. Most participants expressed the belief that benchmarks are likely exaggerations, even benchmarks designed to assure a satisfactory product performance. Only four participants claimed sometimes to consider using more product than the benchmark suggested. Among the frequent points that were expressed:

* People should do research to validate benchmarks for expensive or risky products.

* Acceptance of benchmarks depends on product expertise and familiarity with brand.

* Acceptance of benchmarks depends on trust in the brand.

* Benchmarks are compared across brands as information about value.

The focus groups supported the research premises that benchmarks are an identifiable marketing phenomenon and that consumers process benchmark information in understandable ways.

Study 1

The first study was conducted to assess the main effect of ad benchmarks and to test an experimental procedure. Fifty-one part-time MBA students were randomly assigned to two groups. The samples were appropriate to the study. Their average age was 31; 92% reported owing a car or truck; and 85% reported having played tennis at least occasionally. Each group viewed two ads, separated by a 10-minute distractor task.

The first ad was for Monroe Shocks (with or without the benchmark: "Experts recommend replacing your shocks at 50,000 miles."). The second ad was for a Wilson tennis racket (with or without the benchmark: "Don't wait for your racket to break. Because of racket 'fatigue'--microscopic breakdown of the frame material--experts recommend replacing your racket every year and a half, for people who play regularly."). In one experimental group, the Monroe ad had a benchmark and the Wilson ad did not; in the other group, the reverse. Thus, each subject saw one ad with a benchmark and one ad without. Afterward, subjects were asked to recall several bits of information from the ads, including the advertised product and the benchmarks in the ad ("What did the ad say about how long Monroe shocks (Wilson tennis rackets) should last before needing replacement?"). They were also asked to report their subjective benchmarks ("How long do you personally believe a Monroe shocks (Wilson tennis rackets) should last before needing replacement?").

The results of the benchmark recall and subjective benchmark measures are presented in Table 1. Just over half the subjects accurately recalled the advertised brands (45% for Monroe and 63% for Wilson). More than half who did not recall the specific brand, did recall the product. Recall of the advertised benchmarks was similarly good but not perfect. The average recalled benchmarks were 51,476 miles for Monroe shocks and 1.84 years for Wilson rackets. The better accuracy for the Monroe shocks benchmark may reflect its being a round number, or it may be a result of heavier use in practice. (Monroe appears to use the fifty-thousand-miles benchmark sparingly in advertising; Wilson uses it only through information to retail salespeople.)

In the control conditions, subjects were asked to recall the ad benchmark, even though there was no ad benchmark. I interpreted this measure to reflect a "presumed ad benchmark." For both products, subjects presumed a benchmark that was lower than their subjective benchmarks (63,645 vs. 108,974 for Monroe shocks [t = 8.70, p < .01]; 2.11 vs. 3.61 for Wilson tennis rackets [t = 8.75, p < .01]). These differences support the premise that consumers expect that marketers would present benchmarks that are marketer friendly or unfavorable to consumers.

The results of Study 1 suggest that consumers are influenced by the benchmarks. Subjects who had seen the fifty-thousand-miles benchmark for Monroe shocks reported a subjective benchmark for replacement of 89,338, compared with 108,974 for those who had not seen the marketing benchmark (t = 1.35, p = .09). The subjective benchmark for those who had seen the ad benchmark for Wilson tennis rackets reported a subjective benchmark for replacement of 2.73 years, vs. 3.61 years for those who had not seen the ad benchmark (t = 2.01, p = .03).

Study 2

Study 2 was a replication of the first study, using the same advertising stimuli and same measures. Participants were told they were completing a survey about advertising, then viewed the same two ads, separated by a set of unrelated questions. Each participant saw an ad without a benchmark first and an ad with a benchmark second. (One distinct difference in Study 2 was the inclusion of a measure of advertising skepticism 2 weeks prior to the main data collection; discussed below.) Eighty-one people, undergraduate seniors and part-time MBA students and some adult university staff, completed the online questionnaire and were randomly assigned to one of two conditions. Average age of the sample was 32.4; over 90% reported owning (or expecting to own) a car or truck; and 74% reported playing tennis.

The results of Study 2 are presented in Table 2. As in Study 1, subjects were influenced by the benchmarks. Subjects who had seen the fifty-thousand-miles benchmark for Monroe shocks reported a subjective benchmark for replacement of 71,168, statistically significantly lower than subjects who had not seen the benchmark (85,232; t = 2.11, p = .02). Likewise, for Wilson tennis rackets, the subjective benchmark for those who had seen the ad benchmark was 2.09 years, vs. 2.71 for those who had not seen the ad benchmark (t = 2.48, p = .01). As in Study 1, for both products, the inclusion of a benchmark in the ads reduced people's subjective benchmarks for replacement.

Again, as in Study 1, the presumed benchmarks for ads that did not include a benchmark for both products were lower than subjective benchmarks (66,322 vs. 85,232 for Monroe shocks [t = 3.67, p < .01]; 1.71 and 2.71 for Wilson tennis rackets [t = 5.71, p < .01]), supporting the proposition that consumers expect that marketers would present benchmarks that are marketer friendly.

Study 3

Study 3 replicated the first two studies with key changes. First, different products were used. More significant, the benchmarks were changed in type, direction, and communication mode. Participants in Study 3 were 106 people, about a third adult students and the remainder adult co-workers or friends. The sample was recruited from a part-time MBA class with snowball recruitment by the students. The experimental procedure was identical to Study 1, using different ads. The ads were for Tide powder detergent (with and without the benchmark: "One load--one scoop [3 tablespoons])" and Crest toothpaste, with a picture of a toothbrush loaded with toothpaste (either 15 mm or 30 mm in length--just under a half-brush or just under a full brush in length).

Subjects were more successful than in the previous studies in recalling the advertised brands, 93% for Tide and 85% for Crest than in Study 1--likely due to the more established and popular brands.

The ad benchmark manipulations were slightly different in this study. In the first two studies, both benchmarks related to time to replacement. In Study 3, both the Tide and Crest benchmarks were for amount of product use. Furthermore, the Crest benchmark differed in two other ways: first, it was small vs. large, rather than absent vs. present; second, it was a picture, rather than a statement.

Responses to the perceived ad benchmark measures suggested the manipulations were successful--see Table 3. For Tide, the mean estimated ad benchmarks, in tablespoons, were 2.02 (no ad benchmark condition) and 2.61 (ad benchmark of 3.00); t = 3.57, p < .01. The toothpaste benchmark (length of the squeeze) was estimated by selecting one of 10 lines, which varied from 4 to 40 mm. The means for recalled benchmarks was 18 mm (in the longer squeeze condition) and 11mm (shorter squeeze); t = 5.50, p < .01). Although the conditions differed in the expected directions, the recall of ad benchmarks was not accurate. For Tide, the recalled benchmark was 2.61, lower than 3 tablespoons, as stated in the ad. And for Crest, recalled lengths were 18 and 11 mm, whereas the correct lengths were approximately 30 and 15 mm. Given the familiarity with the tasks--laundry and teeth brushing--and the relative ambiguity of the pictorial benchmarks for Crest, recall of the benchmarks for these products may have been influenced by subjective benchmarks or presumed benchmarks.

Once again, the ad benchmarks influenced subjective benchmarks for both products. Subjects who had seen the ad with the 3 tablespoons benchmark for Tide reported a subjective benchmark of 2.29, statistically greater than the 1.91 reported by those who had not seen the ad benchmark (t = 2.36, p < .02). Likewise, subjective benchmarks for the two Crest ads differed (16.1 for the long squeeze ad vs. 13.3 for the short squeeze ad; t = 2.32, p < .03). Thus, for both products, respondents' subjective benchmarks were influenced in the directions of the ad benchmarks. In the control condition for Tide, unlike in Studies 1 and 2, the presumed ad benchmarks did not differ statistically from subjective benchmarks (2.02 vs. 1.91).

The Role of Ad Skepticism

Obermiller and Spangenberg (1998) proposed the concept of advertising skepticism, a marketplace belief that advertising is untruthful. They reasoned that consumers are socialized to be skeptical of the truth of advertising and that consumers vary on this dimension. I tested a corollary that consumers who are generally skeptical of the truth of advertising are also skeptical specifically of marketing benchmarks in advertising. If so, one should see a moderating effect of ad skepticism on the relationship between ad benchmarks and subjective benchmarks.

In Studies 2 and 3, in a separate data collection, subjects had responded to the 9-item skepticism toward advertising (SKEP) scale (Obermiller and Spangenberg 1998). After correcting for directionality, SKEP scores were computed for each subject, with means of approximately 27 and 29, in the two studies. (Obermiller and Spangenberg [1998] reported means between 28 and 29.5 in their original work.) For subsequent analyses, subjects were combined and divided into "low" and "high" skepticism groups with a median split.

The prediction was an interaction effect: Low skeptical people, who should be more open to influence, should show a larger effect of ad benchmarks; high skeptical people should show relatively less effect of ad benchmarks. Analysis of variance treated subjective benchmark as the dependent measure and the presence/absence of ad benchmark and low/high skepticism group as independent variables. Because the subjective benchmark scales differed across products, the scales were transformed into standardized measures. Thus, the test was conducted on standardized measures of subjective benchmarks for Monroe shocks, Wilson tennis rackets (Study 2), and Tide detergent (Study 3), with ad benchmark and SKEP as independent variables. (1) (No SKEP scores were available from participants in Study 1, and the Crest manipulation was large vs. small, rather than present vs. absent.)

Benchmarks for Monroe shocks and Wilson tennis rackets encouraged people to replace the product sooner than they otherwise expected was appropriate (thus, encouraging smaller numbers); the benchmark for Tide detergent encouraged them to use more of the product (thus, encouraging larger numbers). Data were reverse-coded for Monroe and Wilson, with the result that positive deviations indicated greater influence toward the benchmark (quicker replacement of shocks and tennis rackets, more detergent used); negative deviations indicated less influence (slower replacement of shocks and tennis rackets, less detergent used).

The combined data set had 268 observations, with a 46%/48% split on skepticism (6% missing) and a 51%/49% split on ad benchmarks. The analysis of variance model was significant overall (F = 11.88, p < .01) and showed a main effect of benchmark (F = 32.22, p < .01), no main effect of skepticism (F = .45, p = .50) and a modest interaction effect (F = 2.89, p = .09). The means for standardized subjective benchmark measures were .35 for ads with a benchmark and -.31 for ads without a benchmark.

Figure 1 illustrates the interaction effect. Paired t-tests indicate no difference in subjective benchmarks between high and low skeptics for ads without a benchmark (-.39 and -.26: [pounds sterling] = 1.62, p = .13). When ads had a benchmark, however, subjective benchmarks were higher for low skeptics (.51 vs. .21: t = 3.75, t < .01). Both simple main effects of ad benchmarks were also significant (-.26 vs. .21: t = 5.89, p < .01 for high skepticism; and -.39 vs. .50: t=11.12, p < .01 for low skepticism). Thus, the data suggest that ad benchmarks are influential, regardless of ad skepticism, and that high skepticism provides only limited protection against their effects.

Discussion and Conclusions

Marketing benchmarks are promotional claims about how much of a product to use, how often to replace a product, or how much to spend on a product. The focus groups that began this research suggest that consumers are aware of such benchmarks as a promotion tactic and associate them both with the goal of a satisfactory product experience and with exaggerations intended to increase sales.

Subjective benchmarks are consumer beliefs about how much of a product to use, how often to replace a product, or how much to spend on a product. Experimental subjects showed clear suspicion about marketing benchmarks--in four of five tests, they presumed that ads would provide benchmarks that were more marketer friendly than their subjective benchmarks.

Despite this wariness, all three experiments demonstrated an effect of marketing benchmarks in print ads on subjective benchmarks. On average, the inclusion of benchmarks in ads shifted subjective benchmarks in favor of marketers--more frequent replacement of products and more product use per application.

Notwithstanding the evident suspicion about benchmarks, measures of general SKEP showed only a weak effect on the response to ad benchmarks. On average, people both low and high in ad skepticism were influenced by benchmarks in ads, although high skeptics showed less influence.

Previous research has demonstrated that consumers are suggestible and may buy more, pay more, or use more, when prompted by advertising with product amounts (Wansink 1996; Wansink, Kent, and Hoch 1998). Those results are explained as low-involvement responses or heuristic responses to the marketing cue (either advertising claim or package size). In fact, Wansink, Kent, and Hoch (1998) found that the effect was eliminated when participants were instructed to think about their decisions. That research does not suggest any enduring change in consumer decision making or behavior. Although Study 5 in Wansink (1996) included multiple uses, there were multiple instances in one situation (wash loads at a single laundry session), all of which had the package present. It is the enduring effect, however, that is most critical.

The current research considers a more conscious component of consumer decisions--the subjective beliefs about how much to use, how much to spend and when to replace. For both practitioners and academics, the distinction is important. Although it is significant that marketers can nudge consumers with an ad or a package, it is more dramatic if marketers can effect fundamental changes in consumers' beliefs about products. On the one hand, an ad might increase the number of quarts of oil purchased on one occasion. On the other hand, a change in subjective benchmark might change how frequently oil is purchased for the rest of a person's life. Even something as trivial as toothpaste, for which images indicate at least a 5x amount, would result in a consumer loss of $36 per year (at 80% waste of three $5 tubes per year). (The reader can estimate producer surplus by multiplying by hundreds of millions of consumers.) Marketing benchmarks may have profound economic effects at the macro level.

Future research should explore the role of benchmarks further. In addition to advertising skepticism, we should expect the effect of benchmarks to be moderated by product type, consumer knowledge/expertise, type of claim, and source of the benchmark. In addition, researchers might investigate differences that might explain why some consumers respond to benchmarks as overestimates (too high levels) and some regard them as "minimum levels."

Marketing benchmarks present significant challenges for public policy and marketing ethics. On one hand, the potential financial damage to consumers (and product waste) is great. Many deceptive advertising cases are predicated on a single purchase model, but marketing benchmarks may influence subjective benchmarks that lead to hundreds of repeat purchase mistakes. Furthermore, an interesting possibility looms in the coming "Internet of Things." Appliances and smart speakers, even electronic assistant programs, are likely to be enabled to monitor and anticipate our purchasing needs. To do so, they will need to make decisions about how frequently to replace, how much to use, and how much to spend. Will those benchmarks be carefully decided by their users or defaulted to marketers?

On the other hand, marketing benchmarks may be regarded as puffery, about which consumers can be expected to be skeptical. One might argue that we do not regulate claims about which product is "best," nor do we stipulate maximum prices. We let consumers make their own decisions.

In the middle ground, we should treat marketing benchmarks as we treat more specific claims, with consideration for the long-term financial implications. Benchmarks should not be presented in imperative forms ("Two tablespoons of detergent are required" or "Filters must be replaced after 90 days"), but, rather, as suggestions ("For best results, use..." or "It is recommended to replace after..."). And, marketers should have substantiation for those suggestions--evidence that no more is required for standard performance.

Most people would agree that benchmarks should be monitored and regulated when consumer risk is high--for example, advertised drug dosages and load limits for heavy equipment should be accurate. But should consumers also be protected against such mundane follies as using too much detergent or dental floss? If we consider the lifetime costs of excess, the answer is yes. Marketers must recognize that consumers are influenced by marketing benchmarks, and their decisions to suggest appropriate use levels/replacement schedules/expenditures should be informed by the knowledge that such suggestions are influential. Practitioners must, as always, be constrained by honesty, but regulators should also protect consumers from unsupportable benchmarks.

REFERENCES

Calfee, J. E. and D. J. Ringold. 1994. The Seventy Percent Majority: Enduring Consumer Beliefs about Advertising. Journal of Public Policy and Marketing, 13 (Fall): 228-238.

Darby, Michael R. and Edi Kami. 1973. Free Competition and the Optimal Amount of Fraud. Journal of Law and Economics, 16 (April): 67-88.

Drucker, Peter. 1973. Management: Tasks, Responsibilities, Practices. New York: Harper and Row.

Ford, Gary T, Darlene B. Smith, and John L. Swasy. 1988. An Empirical Test of the Search, Experience and Credence Attributes Framework. In Advances in Consumer Research, vol. 15 (239-243). Provo, UT: Association for Consumer Research.

Ford, Gary T, Darlene B. Smith, and John L. Swasy. 1990. Consumer Skepticism of Advertising Claims: Testing Hypotheses from Economics of Information. Journal of Consumer Research, 16 (March): 433-441.

Kotler, Philip. 1967. Marketing Management: Analysis, Planning and Control. Englewood Cliffs, NJ: Prentice-Hall.

Kotler, Philip and Kevin Keller. 2016. Marketing Management, chap. 13. 15th edition. London: Pearson.

Larkford, Kimberly. 1990. How Much Life Insurance Do You Need?, http://www.kiplinger.com/magazine/archives/how-much-life-insurance-do-you-need.html

Nelson, Phillip. 1970. Information and Consumer Behavior. Journal of Political Economy, 78 (February): 45-57.

Obermiller, Carl and Eric Spangenberg. 1998. Development of a Scale to Measure Consumer Skepticism toward Advertising. Journal of Consumer Psychology, 7 (2): 159-186.

Obermiller, Carl, Eric Spangenberg, and Doug MacLachlan. 2005. Ad Skepticism: The Consequences of Disbelief. Journal of Advertising, 34 (3): 7-18.

Porter, Michael. 1980. Competitive Strategy. New York: The Free Press.

Schmidt, Jeffrey and Richard Spreng. 1996. A Proposed Model of External Consumer Information Search. Journal of Academy of Marketing Science, 24 (3): 246-256.

Tversky, Amos and Daniel Kahneman. 1974. Judgment under Uncertainty: Heuristics and Biases. Science, 185 (4157): 1124-1131.

Wansink, Brian. 1996. Can Package Size Accelerate Usage Volume? Journal of Marketing, 60 (July): 1-14.

Wansink, Brian, Robert Kent, and Stephen Hoch. 1998. An Anchoring and Adjustment Model of Purchase Quantity Decisions. Journal of Marketing Research, XXXV (February): 71-81.

Ziobro, Paul and Serena Ng. 2013. Is Innovation Killing the Soap Business? Wall Street Journal, April 3, available at https://www.wsj.com/articles/SB 10001424127887323916304578400521297972496.

Carl Obermiller (carlo@seattleu.edu) is Professor of Marketing at Seattle University.

(1.) When analyzed separately, only the Tide detergent data showed a statistically significant interaction effect, although results tor all three products showed the same pattern.

DOI: 10.1111/joca.12193
TABLE 1
Ad Benchmark Recall and Subjective Benchmarks, Study 1

               Monroe            Monroe            t-statistic
               Shocks, With      Shocks,
               Benchmark         Without
               (50,000 miles)    Benchmark

Ad benchmark   51,476             63,645           1.35 (p = .09)
recall
Subjective     89,338            108,974           1.47 (p = .07)
benchmark

               Wilson            Wilson
               tennis rackets.   tennis rackets,
               with benchmark    without
               (1.5 years)       benchmark
                                 (1.5 years)

Ad benchmark   1.84              2.11               .96(p = .17)
recall
Subjective     2.73              3.61              2.01 (p = .03)
benchmark

TABLE 2
Ad Benchmark Recall and Subjective Benchmarks, Study 2

               Monroe Shocks,   Monroe Shocks,     t-statistic
               With             Without
               Benchmark        Benchmark
               (50,000 miles)

Ad benchmark   49,603           66,322             3.96 (p = .00)
recall
Subjective     71,168           85,232             2.11 (p = .02)
benchmark

               Wilson tennis    Wilson tennis
               rackets, with    rackets, without
               benchmark        benchmark
               (1.5 years)      (1.5 years)

Ad benchmark   1.57             1.71                .73 (p = .24)
recall
Subjective     2.09             2.71               2.48(p = .01)
benchmark

TABLE 3
Ad Benchmark Recall and Subjective Benchmarks, Study 3

               Tide Detergent,
               With              Tide
               Benchmark         Detergent,
               (3 tablespoons)   Without           t-statistic
                                 Benchmark

Ad benchmark   2.61              2.02              3.57 (p = .00)
recall
Subjective     2.29              1.91              2.36 (p = .02)
benchmark

               Crest             Crest
               toothpaste,       toothpaste,
               with large        with small
               benchmark         benchmark
               (8.0 on           (5.0 on
               10-point scale)   10-point scale)

Ad benchmark   4.55              2.92              5.50 (p = .00)
recall
Subjective     4.02              3.33              2.32 (p = .03)
benchmark
COPYRIGHT 2019 American Council on Consumer Interests
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2019 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Obermiller, Carl
Publication:Journal of Consumer Affairs
Article Type:Report
Geographic Code:1USA
Date:Mar 22, 2019
Words:6066
Previous Article:Identity Threat Assessment and Prediction.
Next Article:Encouraging Tax-Time Savings with a Low-Touch, Large-Scale Intervention: Evidence from the Refund to Savings Experiment.
Topics:

Terms of use | Privacy policy | Copyright © 2022 Farlex, Inc. | Feedback | For webmasters |