2

Alternative Ideas

“Our minds are constantly subjected to influences of which we have no knowledge.”

Walter Dill Scott
The Psychology of Advertising in Theory and Practice (1903: 367)

As we saw in the last chapter, maintaining the myth that advertising is simply persuasive message communication is the ambition of many who work in advertising and marketing. But alongside the majority were a few pioneers who believed otherwise. One of these pioneers was Walter Dill Scott, the man who is credited with having written “the first serious academic study of how advertising works” (Feldwick 2009: 136).

Walter Dill Scott

The place is the Agate Club in Chicago, a prestigious club for businessmen, and the year is 1901. Walter Dill Scott, Assistant Professor of Psychology at North Western University, has been invited to address a group of eminent businessmen connected with the fast-expanding advertising industry. He would almost certainly have included the above extract from his forthcoming book The Psychology of Advertising in Theory and Practice (Scott 1903) in what he said.

From the reports of the meeting the businessmen were delighted with what Scott told them. But it was the turn of the century, and the ad industry was frankly still in its infancy. It was not for another 11 years that a version of the AIDA model would exert its stranglehold on advertising thinking, so possibly these representatives of the embryonic industry were rather more open-minded than their successors.

Scott's view of advertising was that it was a wayward yet seductive sales tool. Extracts of what he said were recorded in a later article by him in the Atlantic Magazine. They include this description:

One man has roughly estimated that seventy-five per cent of all advertisements do not pay; yet the other twenty-five per cent pay so well that there is scarcely a business man who is willing to stand idly by and allow his competitors to do the advertising. (Scott 1904)

Note that advertising was not seen as a marketing tool at this time. The word “marketing” would not be invented for another 5 years, and when it was, it was as an economic function designed to help farmers target their produce more effectively. It certainly had nothing to do with esoteric sales activities such as advertising.

What singles out Scott from everyone else is that he clearly understood that advertising was “manipulating” the minds of consumers. Later in his speech he stated:

As advertisers, all your efforts have been to produce certain effects on the minds of possible customers. Psychology is, broadly speaking, the science of the mind. Art is the doing and science is the understanding how to do, or the explanation of what has been done. If we are able to find and to express the psychological laws upon which the art of advertising is based, we shall have made a distinct advance, for we shall have added the science to the art of advertising. (Scott 1904)

In the same article Scott predicted that “successful advertisers will be … termed psychological advertisers.” He particularly identified how advertising could influence one subconsciously through what he called “suggestion.” In his book he told the story of how a tailor was conducting what he describes as a “vigorous” advertising campaign in Chicago. “I did not suppose that his advertising was having any influence on me,” he wrote, but “Some months after the advertising campaign had begun I went into the tailor's shop and ordered a suit.” Scott then got into conversation with the owner, who asked him if a friend had recommended the shop to him, and he replied that this was the case. It was only afterwards, when he could not recall which friend might have made the recommendation, that he realised he had been influenced by the advertising without his knowledge:

I had seen (the tailor’s) advertisements for months and from them had formed an idea of the shop. Later, I forgot where I had received my information and assumed that I had received it from a friend. (Scott 1903: 176)

Interestingly, he observes that: “I doubt very much if I ever read any of the advertisements further than the display copy” (what we nowadays call the headline). So just the brief copy contained in the headline had been sufficient to cause him to go into the shop and buy his suit.

Later in the same book, in a chapter entitled “The unconscious influence in street railway advertising,” Scott makes perhaps his most famous observation:

One young lady asserted that she had never looked at any of the cards in the cars in which she had been riding for years. When questioned further, it appeared that she knew by heart every advertisement appearing on the line … and that the goods advertised had won her highest esteem. She was not aware of the fact that she had been studying the advertisements, and flatly resented the suggestion that she had been influenced by them. (Scott 1903: 370)

The importance of this quote is that it identifies, presumably deliberately, that advertising can affect many people without them really being aware of it having done so. Possibly that wasn't much of a surprise back in 1903. After all, advertising was mostly restricted to newspapers, hoardings, and ads in tramcars, and I'm sure that these were paid the same sort of scant attention that print ads are nowadays.

As it turned out, in spite of the enthusiastic response Scott received in the Agate Club, the ad industry generally ignored Scott's fascinating speculation about the inadvertent effects of advertising. Instead, they decided in favor of the view of John E. Kennedy, that advertising was “salesmanship in print” (Gunther 1960: 58). And a few years later, in 1910, the industry adopted its own version of the AIDA model, one with the acronym AICA, standing for Attention, Interest, Conviction, and Action (Printers Ink 1910).

The contrast between Scott's observations about the possible subconscious effect of advertising, and AICA's uncompromising categorization of advertising as something that needed to get attention and persuade, could not have been greater. But back in 1910 the ad industry was run largely by ex-sales managers for whom inattention on behalf of the customer was frankly just another word for failure. The nervousness of the industry is illustrated by the writings of Claude Hopkins, head of Lord & Thomas, then the world's most successful ad agency. Hopkins, in an attempt to ape the publication of Fredrick Taylor's best-selling 1917 book Scientific Management, published a book called Scientific Advertising in 1923. Hopkins’ beliefs were that “Advertising is salesmanship … Fine writing is a distinct disadvantage … No one reads ads for amusement … Consider (readers) as prospects standing before you, seeking for information. Give them enough to get action” (Hopkins 1923: 220–222). In other words, when writing an ad, behave exactly as if you were a door-to-door salesman.

So Scott's ideas were forgotten, and attention took center stage for the best part of 60 years, becoming even more important with the arrival of commercial television advertising (Barry & Howard 1990). Of course, once advertising became a national phenomenon in the USA, it wasn't long before cracks started to appear in the neat rational “salesman” explanation of how advertising worked.

One of the first of these cracks appeared in 1962, in the form of an article in the Journal of Advertising. The editorial announcing it stated: “[John] Maloney presents surprising evidence against the conventional view that an advertisement must be believed before it can influence attitudes or behavior” (Maloney 1962: 2). This is indeed what Maloney, a researcher with Leo Burnett, found, and he had evidence to back up his assertion.

It is worth at this point explaining the difference between attitudes and beliefs. Beliefs arise from things we are told about products and services, and so are attributes we ascribe to those products and services. Thus, we may be led to believe by advertising that a Ferrari sports car can do 163 miles per hour, or that Bold washing powder washes whiter than others.

The problem here is that we don't buy anything because of beliefs, because beliefs don't satisfy our needs. We don't buy a Ferrari because it does 163 miles per hour, we buy it because we need a car that is exciting and will impress our girlfriend or our neighbors. And we don't buy Bold because it washes whiter than others; we buy it because we need our clothes to be seen to be clean. These personalized ideas that relate to our wants and needs are attitudes. The important bit to remember is that advertising doesn't affect us by changing beliefs; it only affects us if it changes our attitudes.

So what Maloney was finding was that people's attitudes could be changed by beliefs, even if they didn't believe them. Just think of the implications. For advertising to work without being believed sounds innocent enough; but if people didn't have to believe what advertising told them in order for it to be effective, then the “conviction” part of the AICA model had to be wrong? And if conviction was wrong, AICA was wrong, and advertising had to be something more than simply “salesmanship.”

Two years after Maloney's article, Jack Haskins presented some more controversial evidence of flaws in the idea that advertising was salesmanship. Haskins, the research manager at Ford, discovered that “factual–rational–logical” messages were less effective at changing attitudes than emotional or non-factual messages. He also found that “recall and retention measures seem, at best, irrelevant to … the changing of attitudes and behavior” (Haskins 1964: 7). In other words, rational messages didn't work very well on their own, and remembering what advertising said didn't really matter much. A bit like saying “what the salesman says is irrelevant, it's how he looks that counts.” This, as we shall see, is a lot nearer the truth than anyone ever thought in those days.

Nether Maloney nor Haskins had much success in challenging the attention-driven sales model of advertising, but 3 years later a more serious assault started. This time it was from another psychologist, Herbert Krugman.

Herbert Krugman

Herb Krugman started his career in the psychology branch of the US Army Air Force and spent time on the teaching faculties of Yale, Princeton and Columbia Universities. Unlike Scott, he had worked as an advertising researcher in the ad agency Ted Bates, and also as a market researcher for the research company Marplan. Krugman had built his reputation in public opinion research, and it was almost certainly his hands-on relationship with real people that enabled him to drop his first bombshell. In 1965, writing for the journal Public Opinion Quarterly, he questioned the effectiveness of mass media, describing the impact of television advertising as “limited,” and suggesting that “Television is a medium of low involvement compared with print.” (Krugman 1971: 3).

You can imagine the reaction of the advertising barons of Madison Avenue. This one sentence, suggesting as it did that television as a medium was defective compared with its arch rival, print, must have appeared to them to be the equivalent of a maniac holding a sharpened knife to the throat of the golden goose. But other psychologists shared Krugman's feeling that responses to television advertising were not what you would expect from traditional persuasive communication. In the first place, when people were asked about the influence of TV advertising, they said (rather like Walter Dill Scott's “young lady”) that they “rarely [felt] converted or persuaded” (Krugman 1965: 350). And, in the second place, the very nature of TV advertising content just seemed too trivial and silly to fit in with what was generally seen as persuasive communication. As Krugman put it:

Does this suggest that if television bombards us with enough trivia about a product we may be persuaded to believe it? On the contrary, it suggests that persuasion as such … is not involved at all and it is a mistake to look for it … as a test of advertising's impact. (Krugman 1965: 353)

This article of Krugman's did not excite that much attention amongst the advertising community. After all, he was just a market researcher. But 2 years later Krugman became manager of corporate public opinion at the General Electric Company, the USA's biggest advertiser. Now he worked for a client this meant his ideas had to be taken more seriously.

Krugman's next step was to embark on a series of laboratory experiments into the processes of “looking and thinking and attention and relaxation.” His initial work was carried out using eye movement recording to measure the amount each ad was scanned, the idea being that more scanning equates to more active learning. But he at once encountered a paradox: it was true that on an individual basis the respondents who scanned more recalled more, but on a sample basis it was the ads that were scanned less that were better recalled (Krugman 1968).

To find an explanation for this, Krugman turned to the writings of the celebrated Harvard psychologist, William James (James 1890). James defined two types of attention: voluntary and involuntary. He maintained that voluntary attention could not be sustained for more than a few seconds at a time, and that high levels of attention sustained for longer than this were “a repetition of successive efforts which bring the topic back to mind.” Krugman's experiments suggested that people might actually absorb more from ads that encouraged them to watch in a relaxed, passive state-of-mind (involuntary attention) than from ads that attempted to excite them and make them watch in an active state-of-mind (repeated voluntary attention). This was in line with the findings of a 1964 experiment by Festinger and Macoby. Festinger and Macoby found that if a message in a TV advertisement was played on audio accompanied by an unrelated video, then students learned it better than if it was played accompanied by a video relevant to the message (Festinger & Macoby 1964).

According to Krugman, “Apparently, the distraction of watching something unrelated to the audio message lowered whatever resistance there might have been to the message” (Krugman 1965: 352). In order to investigate this further, Krugman set up an experiment in 1969 in conjunction with the Neuropsychological Laboratory of New York. Its objective was to investigate the difference between the way that the brain deals with TV advertising and press advertising. It is revealing of the attitudes of the ad industry at the time that this was only the second of Krugman's articles to find its way into an advertising research journal (Krugman 1971).

Essentially, what Krugman did was to connect a subject to an electroencephalograph, or EEG machine, and measure the type of brain waves emitted while they were reading a magazine and being shown some TV advertising. He divided the brain waves into three categories: Alpha waves, which indicate relaxation; “fast” (Beta) waves, which indicate alertness, activity and arousal; and “slow” (Delta and Theta) waves, which indicate boredom. Krugman assumed that the difference between emissions of “slow” and “fast” brain waves would signify the level of interest that the subject had in the media being exposed to her. The more fast waves, the more active and attentive was the processing; the more slow waves, the more passive and inattentive was the processing.

The subject used in the experiment was a 23-year-old female secretary. After allowing her to read some magazines, Krugman exposed her to three TV ads, all the time measuring her brain wave emissions. He found little change in the Alpha waves, indicating that the subject was generally relaxed throughout the test. But he did find a dramatic difference in the slow and fast waves emitted during exposure to the two different media. While the magazine was being read, the proportion of fast to slow waves was about 4 : 1, indicating that the subject was alert and interested. But during the TV commercials the proportions were almost equal, suggesting that the subject's level of alertness was far lower than for the magazines. This was excellent support for his view that TV is indeed a medium of low involvement compared with print.

Krugman's experiment had a number of glaring methodologic flaws. First, there was only one subject involved. Second, EEG only measures activity in the neocortex, so any neural activity taking place deeper in the brain will not be registered. Third, the context of the brain wave measurement was skewed: measurement of print advertising covered reading of the whole magazine, not just the ads, whereas the TV commercials were shown in isolation from any programming. Fourth, the subject was sitting in a laboratory.

As I said earlier, advertisers didn't much like Krugman's ideas, and they didn't believe much of what he found in this experiment either. But I have personally repeated Krugman's experiment using a number of subjects, a more accurate measure of attention, a balanced and more representative context, and a more comfortable environment, and the results were exactly the same (Heath 2009). TV is indeed a medium of lower involvement than print.

Krugman's assertion was a categoric refutation of the received wisdom amongst advertisers of the day, which was that TV watching was a high attention activity, certainly higher than reading a print ad. After all, two senses are involved in TV watching – sight and sound – rather than just one, so surely there is twice as much going on to attract the viewer's attention? But Krugman's findings do make common sense. It is relatively easy to pay attention when you are reading text, because you have control of the time you apply to the task – you can scan a page, paying more attention to the parts you are interested in, and less attention to parts you are not interested in. However, with television, the imparting of information depends entirely upon the way in which the advertisement is made. You have no control over it and (unless you have seen the ad before) no real idea of how interested you are likely to be in it. So television becomes a holistic experience: you cannot “scan” the data selectively, and it is therefore quite easy to believe, as James predicted in 1890, that one might slip into a passive, lower attentive state.

The fact that TV was a low involvement medium was not the only dramatic finding from Krugman's experiment. The accepted theory of the day was that interest in TV ads grew with each exposure: the first exposure might be one in which little attention was paid, but once the ad had been seen it was easier for the viewer to engage with the ad and pay attention. To test this hypothesis Krugman played three ads to his subject three times. What he found in every case was that brain wave activity fell with each exposure. In other words, far from paying more attention on repeat viewing, the subject paid less. What is more, this finding seemed to be true regardless of how important the category was to the respondent. Again, this experiment has been repeated, and Krugman's findings have been validated (Hutton et al. 2006).

So what Krugman was finding was that TV advertising was low attention, and that attention diminished with repeat viewing. Interestingly, he did not see this as a shortcoming of television advertising but as a potential strength. He deduced that learning was taking place, but not learning in the traditional sense.

Our initial EEG data [suggest] … television does not appear to be communication as we have known it. Our subject was working to learn something from a print ad, but was passive about television. She was no more trying to learn something from television than she would be trying to learn something from a park landscape while resting on a park bench. Yet television is communication.

What shall we say of it, a communication medium that may effortlessly transmit into storage huge quantities of information not thought about at the time of exposure, but much of it capable of later activation? (Krugman 1971: 8)

Nor did he see in his results a suggestion that this “different” type of learning used in television communication was less effective than press.

As to the question “Which is better?” we are handicapped by our greater familiarity with active and involved types of learning. Our understanding of how passive learning takes place is still deficient and we are not yet sure how to measure its effectiveness in a fair manner.

Although further work with brain waves is indicated, it should be stressed that there is no evidence or speculative inference here to suggest either print or television is “better” than the other, or that fast or slow brain waves are better than the other. Instead, we have a very great need to understand the differences better and perhaps especially to understand better the significance of slow brain waves. (Krugman 1971: 9)

Krugman was primarily a psychologist and researcher and, perhaps because of this, he never sought to design a model that would help advertisers to exploit what he called low involvement. But he gave plenty of clues as to what such a model should be based on. For a start, he discarded the idea that it should be based on purely verbal concepts. The conventional model, he said is the “verbal, look-before-you-leap, reasonable, or ‘rational’ model. … The future of the low-involvement theory is in the non-verbal area” (Krugman 1977: 8). He also understood the importance of repeated exposure: “The theory of low involvement asserts that repetition of exposure has an effect which is not readily apparent until a behavioural trigger comes along” (Krugman 1977: 9). Finally, he understood the critical importance of distinguishing between what he called exposure and perception:

It might be helpful to first look at two related concepts to look and to see. These two variables permit a fourfold classification – i.e. one can look and see, look but not see, neither look nor see and, perhaps, see without looking.

Seeing without looking involves the two phenomena of peripheral vision and conscious vision … looking at an object can only be accomplished within a three-degree arc … You are not especially conscious of that which is peripherally seen. You don't know that you have seen. Later you may deny having seen. Much of what people call subliminal perception is merely peripheral seeing, i.e. seeing without looking at, without being aware that seeing has occurred. (Krugman 1977: 10)

Krugman's ability to conceptualize these ideas so far in advance of others is extraordinary. As we will see in Chapter 6, his notion that you could be influenced by advertising exposed peripherally and not actively processed would not be validated for another 20 years. And his foresightedness went beyond even the realm of processing advertising into the area of decision-making. In his first publication in 1965, he identified the lack of evidence of advertising changing attitudes as being the one major weakness in convincing people of the efficacy of the medium:

The economic impact of TV advertising is substantial and documented. … Only the lack of specific case histories relating advertising to attitudes to sales keeps researchers from concluding that the commercial use of the medium is a success. We are faced with the odd situation of knowing that advertising works but being unable to say much about why. (Krugman 1965: 351)

Then later he suggested what the editor called in his introduction to the article an “arresting” thesis:

as trivia is repeatedly learned and repeatedly forgotten and then repeatedly learned a little more, it is probable that two things will happen: 1) … “overlearning” will move some information … into long term memory systems, and 2) we will permit significant alterations in the structure of our perception of a brand or product [that] fall short of persuasion or attitude change. (Krugman 1965: 353)

In effect what he was suggesting was that advertising worked using some kind of “sleeper” effect. Here's how he sums up his thesis:

I have tried to say that the public lets down its guard to the repetitive commercial use of the television medium and that it easily changes its ways of perceiving products and brands and its purchasing behavior without thinking very much about it at the time of TV exposure or at any time prior to purchase, and without up to then changing verbalised attitudes. (Krugman 1965: 354)

As I said earlier, hidden away as it was in the pages of Public Opinion Quarterly, it was easy for the ad industry to quietly ignore Krugman's fascinating ideas. But in the UK there was another person working on this issue, and finding hard evidence which proved that the persuasion model was flawed. This person was Andrew Ehrenberg.

Andrew Ehrenberg

Andrew Ehrenberg was trained in mathematical statistics, and had been working on consumer panel data. This panel data measured what people claimed to buy, and also recorded what they actually bought. Ehrenberg was finding it very difficult to reconcile these two measures which should, in an ideal world, have been identical.

At the time Ehrenberg was not much interested in advertising, and claimed he didn't really start taking the subject seriously until 1970, when he was appointed by London Business School as Professor of Marketing and Communication, despite in his own words never having read anything on either subject (Ehrenberg 2004). But in 1974 he produced a paper which, even if it didn't overturn the ad industry boat, certainly rocked it quite seriously.

Ehrenberg, like Krugman, believed that advertising was able to work without changing attitudes. His view was that the consumer had extensive experience and in most cases already had extensive knowledge and entrenched attitudes about most of the products that were available. He challenged the traditional notion that advertising could change attitudes and persuade the consumer, asserting instead that “advertising's main role is to reinforce feelings of satisfaction with brands already being used” (Ehrenberg 1974: 33).

At the time Ehrenberg's article was published, persuasion-based hierarchy-of-effects models like AIDA still held sway. Ehrenberg identified four weaknesses in these models:

Ehrenberg did this by using consumer panel data to establish that most markets have few 100% loyal buyers, and that the majority of people buy more than one brand. He showed that brand users held consistently stronger attitudes than non-users, especially evaluative attitudes. But what he could not find was any satisfactory explanation of how attitudes were changed. This led him to question the core assumption within hierarchy-of-effects models that attitude change precedes and drives behavior change. Although Ehrenberg accepted that the traditional idea that “Awareness → Attitudes → Behavior” made intuitive sense, he found many examples of a sequence in which it was the behavior – the purchase of the product itself – which led to “greater awareness of information to which one is normally exposed … and to change in attitude” (Ehrenberg 1974: 30). All this challenged the convention that consumers were rationally persuaded by advertising to change their minds and their brand allegiance.

As an alternative to AIDA Ehrenberg developed a theory he called Awareness–Trial–Reinforcement, or ATR. Note that he was referring to brand awareness here, not ad awareness. The ATR model was popular amongst advertising agencies, at a time when the sales effects of advertising were seen by many as hard to discern even in hindsight, and virtually impossible to predict. Ehrenberg's model suggested that advertising could create, reawaken, or strengthen brand awareness, and could also be one of the factors that facilitated trial purchase. But he also foresaw a role for advertising in converting trialists into satisfied and lasting customers. And he saw repetitive advertising for established brands as primarily defensive, reinforcing already developed repeat buying habits.

In a later article, Barnard and Ehrenberg (1997) refined the ATR model to accommodate what they called “split-loyal” purchasers (those who regularly purchase more than one brand). Here, the role of advertising was to “nudge” these split-loyal purchasers towards a greater purchase proportion of one brand or another (ATR-N).

But even Ehrenberg's ATR-N model generated little enthusiasm amongst marketers. Discussing the popularity of persuasion models in the USA, one of his contemporaries, John Philip Jones, described persuasion as:

[The] conventional view of advertising … which is all but universally believed in the United States and which sees advertising as … a driving force for the engine of demand … capable of increasing sales not only of brands but also of complete product categories. (Jones 1990: 237)

Nor did Jones accept that advertising was incapable of changing attitudes: “[Persuasive advertising] increases people's knowledge and changes people's attitudes [and] is capable of persuading people who had not previously bought a brand to buy it once and then repeatedly” (Jones 1990: 237).

Jones’ view of the way in which advertising achieved this seemed to have changed little from that of the motivational researchers of the 1960s. He described advertising as using creativity to “insinuate” new information into the minds of “apathetic and rather stupid consumers [by] the use of psychological techniques that destroy the consumer's defences; in some cases these techniques are not even perceptible to the conscious mind” (Jones 1990: 237). This contrasted dramatically with Ehrenberg's view of the consumer:

Buyers of frequently bought goods are not ignorant of them. They have extensive usage experience of the products – after all, they buy them frequently. As we have seen earlier, they usually have direct experience of more than one brand, plus indirect word of mouth knowledge of others. The average housewife is far more experienced in buying her normal products than the industrial purchaser buying an atomic power station. (Ehrenberg 1974: 31)

Of course, the crux of the argument between Jones and Ehrenberg was over attitude change. Jones saw advertising working by changing attitudes which led to changing behavior. Ehrenberg, as I said earlier, rejected the idea that attitude change must precede purchase:

It seems to be generally assumed that improving the attitudes of a nonuser towards a brand should make him use the brand, or at least become more predisposed to doing so. But this amounts to assuming that people's attitudes or image of a brand can in fact be readily changed, and that such attitude changes must precede the desired change in behaviour. There is little or no evidence to support these assumptions. (Ehrenberg 1974: 30)

It is worth noting that it is not just the advertising industry that is closed-minded to Krugman and Ehrenberg's ideas. As recently as 1999 a paper on how advertising works published in the world's top marketing journal considered “only theories that adopt an information-processing perspective,” and asserted that “Regardless of their content and the techniques they employ, most [advertising] messages share a common final goal: persuading target consumers to adopt a particular product, service, or idea” (Meyers-Levy & Malaviya 1999: 45).

To illustrate just how narrow minded this information processing persuasion-based perspective is, I'd like you to consider a case study from an award-winning market research paper I wrote in 2008 with Paul Feldwick. This case study concerned a brand of instant noodles.

Telma Noodles Case Study

In 1999, a company called Bestfoods decided to launch a snack food product aimed at teenagers in Israel, called Telma Noodles. The ad agency they commissioned came up with a commercial consisting of a pop song with meaningless gibberish lyrics, accompanying a series of surreally linked and sometimes bizarre scenes. In each scene someone was shown eating the product, but since the lyrics of the song were gibberish the ad contained nothing that could be construed to be a persuasive message about the product.

Research was conducted on this ad amongst teenagers before it was shown. The questions asked were common ones in use at the time, such as “Did this commercial give you enough information about the product?” and “Do you think someone would find this commercial easy to understand?” From these questions, average scores were produced for “ease of understanding,” “believability,” “relevance,” “branding,” and “persuasion.” Since the whole ad was gibberish, the scores on all these dimensions were not surprisingly pretty dreadful. In addition, research showed that the song was widely disliked and detracted from the product. The research agency suggested the ad should not be used and proposed that “Probably a more simplistic route … which emphasises the brand name and benefits clearly would work the best” (Heath & Feldwick 2008: 30).

Heath and Feldwick describe what happened next:

What is unusual about this case is not the research methodology [but] that, for reasons of timing, the advertiser went ahead and ran the ad. The results were exceptional. It became the most liked ad among teenagers … with 93% liking the ad very much – especially the song. Most importantly, the brand took a substantial share of the market. (Heath & Feldwick 2008: 30)

So, rather like the Meerkats, here we have an advertisement that is exceptionally well-liked. The difference, however, is that in the Telma ad there isn't even a message about getting cheap car insurance: there is no message, because the ad is gibberish and totally meaningless. It is a perfect example of Herb Krugman's “trivial and silly” TV advertising. Yet in spite of having no message the ad somehow turned Telma into a huge success.

What the Telma case study does is to bring into sharp focus one of the major schisms that exist between ad agencies and marketers. For most ad agencies Krugman's idea that people don't pay attention to TV ads is heresy, because low attention implies their wonderful work will not be noticed or remembered. So the Telma ad, by being liked and talked about, would be seen by them as a victory for creativity. But for marketers, Ehrenberg's idea that advertising doesn't change attitudes is the worse sacrilege, because no attitude change means there is nothing to measure and no evidence the ad has achieved anything for the brand. The Telma ad, by having no message, would therefore be seen by most marketers as a complete waste of money.

And yet it worked. The brand launch was a success. The ad agency would have been able to use this to argue that recall is the most important thing, and that attitude change didn't matter. You can see now why Ehrenberg's ATR model was so popular with ad agencies; it meant they could stop worrying about getting the message over and spend all their time focusing on creativity that gets talked about.

Except, as Ehrenberg correctly observed, we consumers are not idiots. I'm sure the Israeli consumers didn't go around saying “Did you see that Telma ad the other day, it was complete nonsense, so I think I'll go and try it,” because generally we don't mention ads at all in our day-to-day conversations. Krugman was right in saying we don't pay much attention to TV advertising. In fact, we don't pay much attention to any advertising, and in the next chapter I'll explain why this is.