How many companies are using advertising and brand research methodologies that are positively misleading in their assumptions and their results? The intriguing notion that this could be so was raised last week in a Taylor Nelson AGB seminar, during which two researchers, Jan Hofmeyr and Butch Rice, challenged marketers once again to think more deeply about the concept of loyalty.
Of course, marketers have developed an array of sophisticated ways to measure the impact of their advertising. But according to Rice, joint chairman of the South African firm Research Surveys, most of the results are flawed by a simple oversight of the fact that most consumers have a pre-existing relationship with brands. As a result of this relationship, declares Rice, “you can take exactly the same stimuli – the same advertising, packaging or price – and you’ll get completely different responses”. Consumers who already use your brand, for example, are on average twice as likely to notice the advertising as those who do not.
This means that most quantitative research, which measures scores such as spontaneous and prompted recall, has an in-built bias, which unfairly flatters advertising by big brands and unfairly denigrates ads by smaller ones. Brand managers at the latter will always struggle even to get average recall scores, warns Rice.
This little insight is one example of new light being thrown on marketing by a theory of brand switching called the Conversion Model. The model, invented by Hofmeyr, rejects as one dimensional marketers’ traditional notions of loyalty, which are based on the fit between an offer and a target consumer’s needs and values.
The trouble with this model is that it invites marketers to mentally rank their customers on a loyalty spectrum (from utterly monogamous to determinedly promiscuous), which creates as many problems as it solves.
These problems include the baffling number of satisfied, monogamous customers who suddenly defect; the awful fact that many of the most loyal customers are also the least valuable (they buy your brand 100 per cent of the time, but that’s only because they buy the category so rarely that, by definition, if their last purchase was yours, they’re loyal to you); and the equally disturbing fact that some of your most valuable customers are wildly promiscuous (they buy the category so heavily that even though they only put, say, 20 per cent of their business towards your brand, it still amounts to an awful lot of business).
The Conversion Model adds two new layers of investigation to the needs/values fit. The first is the buyer’s degree of involvement with the brand. This can be straightforwardly financial (for example, if the cost or hassle of switching to a rival is high) or psychological (for example, the emotional pain of leaving). Second, it measures degrees of attraction to alternative brands.
The results create a new set of scores, dividing the total market into users and non-users. Users are split into committed and convertible (that is, those wavering in their loyalty); non-users into available (that is, potentially convertible to your brand) and non-available. And each of these four categories are subdivided into a strong or weak version of commitment or convertibility.
If Hofmeyr and Rice are right, this profile can be used to investigate a growing range of marketing problems, from advertising effects and valuation of brands, through to how to prioritise new product development projects. Go back to advertising, for example, and Rice claims the ad industry’s obsession with cutting through the clutter by making likeable ads is the product of a spurious correlation.
“Liking does not cause noting,” he explains. Not only are existing users more likely to notice ads for brands they already use, they are also more likely to like the ads. And committed users are far more susceptible to the ads than convertible users… and so on.
Indeed, according to Rice, many brands thrive despite, and not because of, the advertising theories behind them. “The fact that so many theories are absolute garbage and have been applied with such great enthusiasm is testimony to the strength of the brand,” he explains.
Here’s another advertising-related consideration: at which point should a brand not bother with advertising? If Rice is correct, the chances for many small brands to make above-the-line advertising work is frighteningly small. The commitment test in one study of a confectionery brand, ranking fourth out of 14 in the market, showed that 61 per cent of consumers were unavailable to it. This means, Rice claims, 61 per cent of the market was likely to ignore, or not even notice, its advertising. Its best chance, he suggests, is to grab the attention of potential converts at the point of sale. On the other hand, because above-the-line advertising works best on the committed user, its main function is to keep existing users committed – and only to a lesser extent can it be used to attract available non-users.
Clearly, for each brand and market, the number of consumers that fall into the categories of committed, convertible or available is crucial. For example, while banks in the US and South Africa earn average committed user scores of 62 per cent and 73 per cent respectively, UK banks score 36 per cent. “The market is very volatile. It is ripe for new entrants,” says Rice.
On the other hand, in other categories, containing products such as tomato sauce – where for some bizarre reason 85 per cent of brand users are committed – would-be new entrants face an uphill struggle.
Take an opposite example: Trevor Richards from Taylor Nelson AGB talks of one packaged goods brand which apparently dominated the market with a 42 per cent share, but which attracted only a minuscule committed figure of seven per cent among its users. It is currently being hit hard by a wave of successful new entrants. Says Rice: “Before going for a market, it’s worth having a look at its commitment structure.”
This month, Hofmeyr and Rice begin tests which adapt Conversion Model research techniques to interactive media, such as the telephone and the Internet. Theoretically it will enable marketers to tailor their response, in real-time, according to whether the person at the other end of the line is an entrenched user, convertible away from the brand, strongly available or strongly unavailable. The results of the trial will help persuade marketers whether Rice’s analysis is correct or whether they should continue with their existing evaluation methods.