There was an item on Marian Finucane’s radio show today about loyalty cards. I was disappointed that the knowledge of all of the people on the show about how loyalty cards actually work in economic terms, and how the data collected is actually used was very weak indeed.
First thing: economics. Loyalty cards are primarily about economics. They are about maximizing profits for the shop (not that there’s anything wrong with that). The way they do that is primarily through discriminatory pricing. Discriminatory pricing means charging different prices to different people for the same product. Discriminatory pricing is at its most blatant in the travel industry, where the same flight or hotel bed costs a different amount, depending on when you book it, and whether you book it as part of a package. But you also see it in other places – for example, a can of Coke is likely to cost more in a centre city location than it costs in a store in the same chain in an out-of-town superstore -.
If you can figure out a way to charge different people different prices for the same product, you are on to a winner. You no longer need to worry about upsetting your price-sensitive customers with price increases. You no longer need to worry about doing major damage to your margins by cutting your price all at once. Instead, you can be subtle with your pricing – charge as muchas what each customers can afford and is prepared to pay –
Tesco uses its loyalty card to do this, by offering discounts to people who care about the price of groceries and bother to use the card, versus people who really don’t give a damn, and don’t sign up or never use their card. Marian Finucane said that the vouchers that you get are just ‘junk’, but that’s easy for someone earning EUR 300k a year to say; The money-off vouchers that Tesco sends out can easily give you 1 or 2 percent off your weekly shop (from my own observations). If your shopping for a big family and aren’t wealthy, that’s a worthwhile saving. It’s also quite a substantial discount for Tesco to give, since their margins are relatively small.
The reason Tesco doesn’t mind giving these discounts is because it doesn’t have to give them across the board – it only has to give them to people who bother to sign up and use the system and who actually take the time and trouble to bring the vouchers to the shop. This means that Tesco doesn’t have to cut its profits significantly to be competitive for these cost-conscious customers. It also avoids the risk of starting a needless price war with a competitor.
In practice, the fact that loyalty card schemes are targeted in this way means that the people involved are more likely to be from amongst the poor. And this is one of the possible problems with successful loyalty card schemes – they target the poor -. We should think hard about whether this is really appropriate, because it means that the decision to use a loyalty card is based on economic necessity, not on completely free choice.
Another thing that is little understood is how Tesco makes money out of the scheme. One way is through offering manufacturers of goods (like Procter and Gamble and Unilever) the opportunity to target discounts on particular products to particular groups of consumers. (This isn’t necessarily a bad thing, but it is good to understand how exactly the economic model for loyalty cards works.)
There were dark murmerings about how the company behind Tesco’s scheme sells the data they collect. The representative from the Irish Data Protection Commissioner’s office said that this was illegal, and the UK-based author of a book about the Tesco system seemed to be the same. However, they are mistaken. DunnHumby, the company behind the Tesco scheme, does in fact sell the scheme, but uses a technical means to avoid breaking the Data Protection laws. Rather than selling the personal information, DunnHumby ‘postcodes’ the data. This means that they sell information about each postcode, rather than about the individual customers. Each postcode in the UK is likely to hold around 20 or 30 households, and people within a postcode tend to be similar, so the information sold is of pretty high quality.
The information could be used, as suggested, to target a health promotion campaign at people in postcodes where few fruit or vegetables are consumed; equally, it could be used to target junk products and take-away food at the people most likely to be interested in them.
It’s worth reading this interview with Clive Humby.