Run with Foxes. Paul Dervan

and pack size matter too. My daughter loves green tea flavour ice cream – she doesn’t care about the brand. I prefer Pepsi to Coke, but size (500ml) matters more than brand. Because I buy it most often doesn’t mean it’s the brand I like most. Another brand might be my favourite, or I’m just the shopper and not the consumer. Some people decide mainly on price – brand doesn’t really matter to them.”15

      This leads us to the famous law-like principle known as double jeopardy. We don’t have many laws in marketing, but this is perhaps as close as it gets.

      I first discovered double jeopardy in 2010, when Professor Byron Sharp’s newly published How Brands Grow caused me some mild panic. Marketers have known about double jeopardy since 1969, and I was just learning about it in 2010. Clearly, I was late to the game. I blame my dad, of course. He was responsible for the reading list in my early years. Surely among all his correlation-causation bed time stories, he might have chucked in a few nuggets on the NBD-Dirichlet theory of repeat purchases. By 2010, I had been making marketing decisions for a good decade. And if you’d asked me about double jeopardy, my best guess would have been that it was some sort of legal loophole where you cannot be tried again for murder if found not guilty the first-time round.

      Broadly, this principle tells us that brands in a category tend to have similar levels of loyalty. Behavioural loyalty. What this means in plain English, is that Starbucks customers visit Starbucks approximately as many times as Costa Coffee customers visit Costa. And Sure deodorant customers buy similar amounts of Sure as Nivea customers buy Nivea.

      I’m simplifying it a little, I suspect, but the practical implications of double jeopardy are easy enough to understand. While repeat sales from existing customers are critical for most brands, what double jeopardy tells us is that we cannot vastly improve how much our customers buy from us, at least compared to the category average. The amount your customers will spend with you has a ceiling. In fact, you can often forecast it with reasonable accuracy.

      If the average customer in our category buys their favourite brand, say, 12 times a year, this is approximately what we can hope for. Maybe we can improve on this. Maybe 14 purchases. Perhaps even 16. But we are not going to double the category average. Our customers won’t be buying our brand 24 times, no matter how customer-focused we are. It’s just not happening. This is a characteristic of many markets. Not much we can do about it apparently. If we are writing up marketing plans that are dependent on far higher levels of purchase loyalty than the category norm, we may be setting ourselves up for failure.

      If this is depressing – unless you are the market leader – it gets worse. It’s called double jeopardy for a reason. Smaller brands get a double whammy. They are hurt twice. So, while the number of times customers buy the various brands tends to be about the same – the larger brands actually get bought slightly more often. So, if the brand leader is getting 12 purchases a year, chances are, you’re not even notching up 12. Sorry.