The unbearable lightness of buying, as told by an old jar of pesto
That doesn’t look normal…
I was holding a jar of pesto: ‘Best before 19 March 2012.’ After unscrewing the cap, it took only one look to realise this stuff would end up in the bin. I shouldn’t have bothered to look. After all, it was already 24 March… 2018.
Hundreds of products must have passed through that fridge. Yet over the last six years, this little jar has spent its time safely tucked away in the corner, blocked by other, more frequently used products. I had completely forgotten about it. Surely, my purchase will have been of little value to this pesto brand? Let’s put that assumption to the test.
What does normal look like?
Below is an example from my work in a non-alcoholic beverages category to illustrate what a typical customer base looks like for a popular brand. People bought this brand over the course of a year, although, it turns out, at very different rates; some buying far more frequently than others.
Figure 1: A lot do a little, and a little do a lot
The brand is well known, yet almost three quarters of its buyers bought it five times or less, together realising nearly 30% of annual sales. This seems surprising, yet marketing science says that it is quite normal. The pattern is closely predicted by a statistical curve, the negative binomial distribution or NBD. Every brand’s customer base is like this, as Ehrenberg (inter alia) has pointed out.
This pattern in buying behaviour is so regular that it can be successfully modelled, and the predicted values used to understand and benchmark actual or future brand performance measures. The pattern underpins much of the evidence presented in Byron Sharp’s 2010 book How Brands Grow, and is incorporated in the underlying theory of the NBD-Dirichlet model, labelled as one of marketing science’s greatest achievements.
I will not dive into the details of the model here, but one of its most intriguing and counterintuitive assumptions is that markets are stationary. Let’s examine what this pattern can tell us when we look at multiple years of buying, and investigate a well-known brand that grew substantially over that time period.
Read the full article on Mumbrella.