Comment: Engagement strategy trumps loyalty programme

We just need a loyalty programme, don't we?

Last year UK retailer Marks & Spencer (M&S) introduced its Sparks card that offers access to seasonal previews, sales and events. Despite an initial sign up of 3.3 million customers, this is a pretty traditional loyalty offering and one that comes very late to the market.

In contrast to the offering from M&S, UK online fashion retailer Asos launched its loyalty programme last year on a different model. Whilst being able to accrue five points for every pound spent with Asos, customers can redeem these at other retailers such as Krispy Kreme, Spotify and STA Travel – retailers that share the same demographic. In addition to this, Asos encourages customers to post pictures of themselves modelling Asos clothing on Instagram to attract further points. These photos submitted by customers are then used on Asos Marketplace to promote the products.

These two stories illustrate the loyalty challenge: in a connected world loyalty comes from an engagement strategy, not from one with a membership model based on Tesco's Clubcard.

So loyalty done well works – but isn't that all about data analytics?

In my view the topic of customer loyalty generates far more heat than light. We all know that customers today are not as loyal as those in previous generations. They won't stick to one supplier and are happy to have multiple relationships.

The common response to this is a loyalty programme and many commentators point to Tesco, Target and others as exemplars in how to retain customers and drive greater basket value through understanding their purchasing history. Yet, whilst it would seem that almost 92% of the UK's adult population are registered with a customer loyalty programme – and with three programmes being the average – around 20% have not redeemed any of their programme benefits.

The loyalty card/programme was not a new idea. The Co-op's membership programme is a very early example of a scheme to create value through regular custom and close association; the 'Green Shield Stamps' phenomenon encouraged repeat purchase in stores that offered them in the 1960s and 1970s and, in high-end retailers, 'the account customer' has been offered additional benefits and exclusives for decades.

What made the loyalty card as developed by Target and Tesco different was that it held data on customer buying patterns on a mammoth scale; and then that techniques applied by the high-end retailer to a few customers could be applied en masse across a customer base of millions. This was, however, a one-way relationship with their customers: the retailer analysed the data and from this came to conclusions on the propensity for a historical pattern of purchase to repeat in the future. What the card didn't offer was the ability to listen to the voice of the customer. It wasn't so much customer relationship management as customer data management and begs the question about whether algorithms can ever accurately predict behaviour.

The big data paradox – did Tesco just get lucky?

When we were researching 'Leading Digital Strategy' we came across a piece from Radio 4 presenter and statistician Tim Harford wherein he analysed whether or not this propensity modelling approach added any real value. His conclusions from talking to world-leading statisticians and analysts was that Target and Tesco gained less through the quality of their algorithms and more because of the vast amounts of data they could harvest. Harford concluded that, with a customer base of scale, the law of averages is just as likely to be the cause of success as any targeting strategy based on historical purchase. As highlighted in his Financial Times article, he concluded that, when you are the size of Target or Tesco and you get lucky some of the time, the small percentage in uplift will create a significant level of gain.

His conclusion, that big data on its own was as likely to generate the wrong insight on behaviour as the right one is, in our experience of retailing and subscription businesses, borne out by the data. He argues for the adoption of a 'deep data' approach – one where businesses learn to apply the principles of ethnography to the generation of data and combine the quantitative with the qualitative. Without this, all any big data set can give you is a number and the opportunity to guess why. The law of averages suggests that, over time, you are likely to get as many guesses wrong as right. Therefore, relying on doing a better job at guessing than your competitors is a very risky strategy.

Using this logic, we have developed our distinctive #'Customer to Action®' and 'Engagement to Action' processes that have driven considerable success in driving growth.

Customer engagement – the essentials

The Asos model is an interesting one. It exploits the connectivity of online retailing to generate a conversation with the customer through a range of channels and works in a way that, rather similar to the co-operative principles, they become part of the enterprise. When we are part of something we feel a sense of ownership. With this comes loyalty.

The loyalty offer starts with a deep understanding of the customer. To develop this effectively requires your online channel to become as much a receiver as a transmitter. This means that you learn to utilise the channel to listen; and that means listening to understand, not just listening to reply. This insight, done well, is the deep data you will need to build the relationship you want with your customers.

This approach should then be replicated throughout your customer engagement online. In our work with membership and subscription businesses, we have discovered that a continuous conversation with loyal customers helps businesses define not just what is missing from their current engagement and proposition, but how to develop the proposition such that it works better and retains engagement with the brand.

A loyalty card and the associated programme will give you nothing but administration costs, a technology headache and a lower margin (although perhaps not as low as it could be as you bank on low levels of redemption). If you are lucky enough to have millions of customers, then the data processing capability you require will, at best, generate a small return as you get lucky with some of your targeting. In the end you risk burning significant investment for an uncertain future.

A deliberate programme of engagement, however, could be transformational.



James Hammersley is the co-author of Leading Digital Strategy a guide to eCommerce strategy published by Kogan Page. He is a founding partner and director of Good Growth, a digital change consultancy which has worked with organisations such as The Economist, The Co-operative, O2 and Manchester United.