Once upon a time, marketing theory advocated targeting a brand’s loyalists (ie more frequent buyers, otherwise known as a brand’s fans). The popularity of this idea, especially at board level, meant every marketer was under pressure to launch a loyalty scheme. All brands have some customers who are worth much more than others, but the idea that these buyers are an easy source of sales growth is flawed.
In the last years of World War Two the allied generals might have preferred not to have had to invade mainland Europe, with their naval and air superiority they’d have preferred to fight out at sea, on islands, or open plains, but not in cities. Unfortunately they had a war to win.
Once upon a time, marketing theory advocated targeting a brand’s loyalists (ie more frequent buyers, otherwise known as a brand’s fans). The popularity of this idea, especially at board level, meant every marketer was under pressure to launch a loyalty scheme. I wrote about this, calling it the “heavy buyer fallacy”. All brands have some customers who are worth much more than others, but the idea that these buyers are an easy source of sales growth is flawed. How can they be? They already buy a lot, and there aren’t many of them.
The book “How Brands Grow” dented enthusiasm for targeting loyalists. The evidence showed, and explained why:
The implications for media strategy have been supported by recent empirical research:
eg best paper in the Journal of Advertising Research 2020: “after carefully accounting for causality, television campaigns are shown to be more effective for lighter brand users” (Assael et al 2021).
And new research has shown that “flighting strategies are inferior to pulsing-with-maintenance and to even-spending scenarios” (Gijsenberg & Nijs 2019), because these latter advertising schedules deliver more reach per $ in any time period.
Market share growth requires nudging the propensities (loyalties) of all category buyers.
It’s why, some years ago now, I delivered an upbeat assessment of the progress of marketing science to a group of academics and CEOs at the Wharton School, saying that one of marketing strategy’s most fundamental questions has been solved, i.e. “who to target?” Market share growth requires nudging the propensities (loyalties) of all category buyers.
“Target the market” started gaining traction as a strategic mantra.
Yet the idea of targeting only a proportion of buyers is one of those zombie ideas that refuse to die, for a number of reasons:
Plus there is money to be made for media owners by selling advertisers a new way of reaching special segments of the buyer base.
The latest iteration of this old idea is to skew advertising spend towards buyers who are most likely to respond. ‘Propensity modelling’ of some sort (see Fader & Hardie 2009, sometimes also called “look a like” modelling) is sold as a way to identify those who are more “movable”. Logically, tautologically, if your advertising only reaches the people who are most likely to be moved by it then it will be most efficient. However, efficiency isn’t the same as effectiveness, and there are many traps in business where long-term gains are eroded at the expense of short-term wins. I’m going to explain why this strategy is just as dangerous as falling for the ‘heavy buyer fallacy’.
Before I do, there are a couple of practical limitations which themselves skewer the idea. Firstly, it’s easy to fool yourself (or your client) with modelling that misattributes advertising effect as “change” or “impact” when many of the buyers were people who were going to buy the brand anyway. There’s a long history of analysts not understanding ‘regression to the mean’, and misinterpreting stochastic wobble as a caused reaction (eg Balding & Rubinson 1996). Secondly, there is a practical problem that bespoke targeted media strategies are expensive.
Let’s put these practical problems aside for a moment; let’s fantasise that we can easily and correctly identify those who truly are most likely to change their buying if they receive an advertising exposure from us, and that we can cheaply and accurately deliver that exposure. So our advertising will be more efficient, indeed the tighter we go with our targeting the more efficient it will be. At an extreme we could send only a single exposure to the sole person most likely to react to it (delivering the highest ROI possible). Or we could do some optimisation modelling to play around with deciding how much to spend to obtain a particular ROI.
While this engineering sounds very ‘scientific’, it isn’t, because marketing science has shown us how buying propensities are distributed across a brand’s buyer base (see Fader et al 2014), and what the new distribution of buying propensities looks like when a brand grows or declines. The implications of these discoveries are stark. For market share growth, a brand needs to nudge propensities right across the market, particularly those buyers with very low propensity to buy the brand, because that’s most consumers. To prevent market share decline, a brand needs to maintain propensities right across the market, particularly amongst those with a very low propensity to buy the brand, again because this is most consumers, i.e. for most consumers the brand is a tiny part of their lives and easily forgotten or ignored.
Ehrenberg-Bass Institute research shows that 25%+ of this year’s sales will (need to) come from buyers who didn’t buy the brand last year. That’s just to maintain market share. The percentage needs to be much higher to grow market share. This is because, even for large (top 10) consumer brands, it’s typical for almost half of their buyers to buy them only once in five years, and for 75% of their buyer base to buy them less often than once a year. These ultra-light buyers don’t sound attractive, yet they deliver almost half of a brand’s sales. And growth isn’t possible without gaining more from these buyers, and gaining more like them.
It also means it’s terribly easy to misclassify consumers when using short time windows. Even a 12-month window will misclassify many brand buyers as being a non-buyer, and will misclassify many long-term buyers as being newly recruited.
Given these facts, a strategy to focus on any portion of category buyers1 isn’t a growth strategy. Indeed, it’s downright dangerous, because it’s hardly a maintenance strategy either. It’s a strategy that trades future sales for efficiency today.
Imagine that you are an industrial sales representative, selling something like earth moving equipment or cloud computing. You’re told that you need to make ten sales to make this quarter’s sales target. Fine, you say to yourself, there’s five customers who are very likely to place an order this quarter, a visit to each of these and that should deliver 3-4 sales; very efficient. You could stop here, it would be the least effort (highest ROI) but you’d miss your sales target by a long shot. So next there are another 10 prospects who will be more work, lots more visits to try and get them over the line, and you might generate 3-4 sales from them. So, if these were to come off, you’d still be shy of your sales target by 2-4. Then there are another 20 prospects who are each rather unlikely to buy, but without visiting all these firms you’re pretty well guaranteed to not hit your sales target.
Or, consider another analogy…in World War Two the allied generals might have preferred not to have had to invade mainland Europe, with their naval and air superiority they’d have preferred to fight out at sea, on islands, or open plains, but not in cities. Unfortunately they had a war to win.
“Activations”, i.e. attempts to catch a share of those buyers who are about to fall (i.e. buy from the category), benefit from propensity-based targeting, that is, reaching those people who are just about to make a purchase (the old, ‘the right message, to the right audience, at the right time’). That’s why most “performance marketing” is done in-store or with search advertising. But “activations” aren’t enough to maintain our brand, let alone deliver growth. Most of our efforts to catch buyers who are about to buy go unnoticed by buyers whose mental availability favours other brands. That’s the role of advertising, to give our physical availability and our activations a greater chance of winning sales, by building mental availability long before the purchase.
It sounds attractive to target heavier category buyers, particularly those heavy category buyers who rarely purchase your brand. Research shows that consumers rarely attitudinally reject brands (Romaniuk et al 2012). It is more the case that they just don’t buy a brand because of lack of mental and physical availability. So these would appear to be the golden households, with the best likelihood of being nudged by your brand’s advertising. Indeed, but before rushing down the ROI dead-end again it’s worth remembering the Natural Monopoly law (see Dawes, 2020). This describes how the customer bases of large brands aren’t just quantitatively different (i.e. larger, more buyers) but also qualitatively different. Larger brands have a greater percentage of light category buyers making up their buyer base. Or put another way, small brands skew towards heavier category buyers. So a strategy to target heavier category buyers is a bit like saying “let’s become a small brand again”, which is what may very well happen as a result if you were to pursue that approach.
New Ehrenberg-Bass Institute research shows that a brand’s sales growth is strongly influenced by category growth, and that this is especially true for the largest brands in any category (Tanusondjaja et al 2021). And category growth is strongly dependent on penetration growth, especially true for newer categories. So to stimulate this sort of growth it’s necessary to reach very light buyers of the category and even people who haven’t (yet) even bought the category. Which is the complete opposite of targeting the easiest buyers.
So, over and over, the evidence is that targeting the soft, easier segment of the market will sacrifice your future. As I’ve written before, “chasing ROI can send you broke” (Sharp 2013 p.97). The companies with stellar stock market valuations are those that investors think/hope have a wonderful future, and it’s that assessment of the firm’s future that matters so much more than getting a cheap sale today. Please keep this in mind when the next consultant pitches an advertising strategy that sacrifices reach. Remember the salt analogy: a little bit of targeting is like adding salt in cooking, it quickly becomes too much.
By all means, do seek efficiencies, but beware the illusory efficiency of tighter targeting.
______________________________________________________________________________
1 One group strongly advocating propensity based targeting of advertising admits that for most brands their moveable segment represents “a single digit percentage of the overall population”!
______________________________________________________________________________
Ehrenberg-Bass Institute scientists:
Others:
Can’t find what you are looking for? or have some feedback about the site? Contact Us
Now available as an eBook exclusively to Apple iBooks
The Ehrenberg-Bass Institute for Marketing Science is the world’s largest centre for research into marketing. Our team of market research experts can help you grow your brand and develop a culture of evidence-based marketing.
