Share of wallet isn’t enough

In a recent Harvard Business Review article TIm Keiningham et al (Oct 2011) argue that managers should pay attention to “share of wallet”. To grow brands should aim to improve their share of wallet rank.

To do this you obviously have to get customers who currently give you a very small share of their purchasing to give you a greater share – it’s logically impossible to get much more share out of customers who already give you near 100%.

So Tim Keiningham et al have discovered the importance of light customers. Great.

Unfortunately, in their article they then make an unsupported assertion that the way to improve a brand’s share of wallet metric (and hence market share) is to survey customers on their motivations for buying each brand and then whatever it is that they like about a competitor should be improved in your brand. This ignores the very weak link between claimed motivations and behaviour. But is an unsurprising recommendation from someone who works for a market research agency.

Like Reichheld and Sasser (see retention profit myth) they also imply that improving loyalty metrics is easy – just ask people what they are looking for, provide it, and then your share of wallet metric will jump.

They provide (only) a hypothetical example of a supermarket. So let’s look at real data on supermarket loyalty. This is Kantar Worldpanel data (2006) on UK supermarkets (a very vibrant and competitive grocery market), market share is in the left column, penetration next, and share of purchases in the right:



Like all loyalty metrics, share of purchases rises with penetration and market share, in accordance with the Double Jeopardy law. As expected, there is much greater variation in penetration than in the loyalty metric.

In the HBR article’s fictional example the supermarket achieves a 7 percentage point gain in share of wallet (at some unknown cost), the implication is that this is an easy task. But this would be equivalent of Sainsbury’s doubling its market share – that’s a Herculean task!  And, very importantly, Double Jeopardy shows us that Sainsbury can’t do this without also increasing its penetration from an annual 64% to something nearer 80% – in other words it has to gain more customers.

That means the supermarket has to increase its reach (in space or time), e.g. more stores, longer hours.  This vital message is missing from the HBR article.

Professor Byron Sharp

Oct 2011

12 thoughts on “Share of wallet isn’t enough

      • SCR = share of category requirements i.e. of all your buying of the category how much goes to brand X. e.g. if I buy toothpaste 5 times a year and Colgate 3 times, then Colgate satisfies 60% (3/5) of my category requirements.

      • Sorry, I have little backgroun in FMCC. How is MS different from SCR?

      • Market Share is a brand level metric. A brand’s share of total sales.

        SCR can be calculated for every consumer. And then the brand’s average is usually reported – for those that bought the brand at all during the period.

        So a brand with 5% market share could still hypothetically have a SCR amongst its buyers as high as 100%.

      • How would you find out, what is the proper “given period” for a given category? If too short (eg yesterday, last hour), it´s almost random. Isn´t it?

  1. Yes, even if brands did manage to grow through increasing loyalty (which very few have been empirically shown to have) it also presupposes a world where brands have recognisable differences. We know that any MEANINGFUL differentiation is only temporary until copied. We tried to look at this in a past client side life to see how we could do exactly this and the only significant attribute that differentiated the leading brand in our market from our #2 brand was “is the best”. Marketing, like life unfortunately, is not that simple.

  2. Nice to read… have to get customers who currently give you a very small share of their purchasing to give you a greater share – it’s logically impossible to get much more from already loyal customers

Please comment on this article if you wish

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s