Calculating cashback, assessing airmiles & rating rewards

Next week we’re launching our new Credit Card Product Ratings. It’s our first foray into rating banking products (though they’ve featured in our Customer Experience Ratings for years) and one I’m quite excited about. As the creator and architect of the product ratings (Ed: Surely a team effort?), I think credit cards have been the most challenging area. Whilst the nuances, intricacies, and subtleties of insurance exclusions are problematic from a ratings perspective, for me credit cards have been a step beyond even that.

When is a credit card not a credit card?

Whilst some ratings agencies simply rate ‘credit cards’ as a general category, we felt that this was intensely problematic. Comparing 0% balance transfer cards with cashback cards seemed bizarre – they’re designed for completely different tasks. It’s the equivalent of comparing contents insurance with buildings insurance – yes, they’re both ‘home insurance’ and there is some crossover, but they’re fundamentally incomparable products.

I could spend time talking about how we decided the categories for our ratings but that isn’t what this article is about. This article is about one type of credit cards – rewards. And by rewards (for this article only) I’m talking about rewards, cashback, and airmiles cards.

Working out rewards

After we had collected all the product information, we started thinking about how we would calculate the amount in pounds and pence that a customer could get from reward cards. Even for someone with all the product information, a PhD in neuroscience (you can call me Dr Tom), and a history of lecturing statistics, I found it extremely difficult to estimate the return – god knows how customers are supposed to do this! But I’ll come to that later.

Commonly cashback cards offer better returns for shopping with the provider than with other retailers. For example, Asda’s Cashback Plus card offers 2% cashback on any spending on their website or stores, but only 0.2% everywhere else. However, it also offers 2% cashback at Asda petrol stations, as well as a whopping 10% back on their insurance products.

To try and work out the possible return for customers as accurately as possible, we used the public data provided by the Office for National Statistics on national spending patterns. This data allowed us to see the average spend on insurance products, petrol, food shopping, etc. And this allowed us to be realistic in the amount of cashback returned for customers.

However, the problem with using this method is it makes two important assumptions. Firstly, that cardholders are typical spenders, and that the amount they spend on petrol, for example, will only be at the place they would get the most cashback. The former makes our estimate conservative, whereas the latter makes it generous. Being conservative means our estimate represents the minimum a customer could get back. The second assumption is slightly more problematic as it suggests consumers’ purchase behaviours are logical, whereas they’re typically not. However, being wrong due to a customer not using their cards efficiently as they should is something we feel comfortable enough with.

We’ve then applied this math to airmiles and reward cards too. E.g. if Amex gives you more Avios points for spending abroad – such as with the Preferred Rewards Gold card – then we used this line of the ONS data to assume that typical amount of spending.

Of course calculating the amount of points you earn is typically only half of the equation. The other challenge is working out how much they’re worth. When it comes to reward schemes like airmiles, the value can vary considerably. On this matter, we’ve gone with the straight forward cash in value – which again, is fairly conservative, but gives you a realistic picture of the minimum typical value you might get out of a card in pounds and pence.

Does this math matter?

And after all that, one question remains – is the potential return for a customer what they would base their decision on when taking out a credit card?

Honestly - probably not. It’s a factor, but with calculations being so difficult (for airmiles and rewards in particular) other factors must have a bigger effect. Relative importance (the value of that reward) to the customer plays a much bigger role in my opinion. If I shop at Very.co.uk a lot, for example, then I might choose their credit card over M&S’, despite M&S potentially having a greater return in pounds on my total spend.

There’s no perfect way of assessing the value of reward cards. And from the card providers’ perspective, that’s part of their beauty. The game is trying to give customers the perception of value – whilst not actually giving them an awful lot.

So our ratings are designed to help people get a quick idea of what cards offer better value. And in the complex world or rewards and cashback, we think what we’ve come up with achieves that goal. Let us know what you think.