How to best use the data you’ve got

When it comes to gaining customer insight, it’s easy to assume that online retailers have it easy compared with their bricks-and-mortar counterparts. Every sale they make can be clearly associated to an individual shopper. Details of a transaction are neatly linked to that customer’s record − containing, at the very least, their full address details and often further demographic information − as well as to their complete buying history. But things aren’t as difficult for store retailers as they seem.

Filling in the gaps

It’s true a store purchase can be pretty anonymous, especially if a customer pays by cash. In a recent survey of UK retailers conducted by Hitachi Consulting − including Argos, Bhs, Debenhams, Halfords, Thorntons and Waitrose − almost two-thirds of respondents (65%) said customer data represented the biggest ‘information gap’ in their corporate information systems. However, only 30% reported gaps in online activity data and just 25% cited product, supplier and finance data as areas of concern.

But it’s not as bad as it looks. For some large retailers, a loyalty card scheme can offer the level of insight retailers need.

Peter Bull, a retail expert at management consultancy firm PA Consulting, says: “You know where customers live, you can analyse what they buy on a regular basis and identify changes to their buying habits.” But getting a loyalty card scheme up and running is too costly for many.

“What many retailers fail to recognise is that there’s still a great deal of customer information to be culled from electronic point of sale (EPoS) data,” says Hitachi Consulting director of retail solutions Chris Gates. That data, he says, can enable a retailer to analyse what products are bought together within an individual basket, judge the success of promotions or in-store merchandising, and spot patterns in customer preferences.

By examining baskets according to average spend, type and number of items, as well as the time of a visit, retailers can build up a picture of what type of shopping trips their customers make at different times of the day, week or year.   

Meanwhile, average spend per basket can be a rich source of information about a particular store’s performance against others in the same chain, as well as its appeal for demographic groups who live locally. “One of the propositions we’re currently working on with our technology partners is helping retailers without loyalty card programmes to deduce patterns from raw EPoS data and match that data to customer segments,” says Gates. “You can go back historically and not only see total sales, but also total sales in each customer segment, or how demand changes among each segment in line with changes in price or product mix,” he says.  

EPoS data has a valuable role to play in assortment planning by helping retailers pick products that they have good reason to believe will appeal to the customer base. “One retailer that I worked with saw a 2% uplift in like-for-like sales, with a 4% reduction in stock, over a six-month period by using EPoS data in this way,” he says. It can also be used to make decisions about distribution, so that inventory is allocated across different channels and stores according to what the retailer knows about the customers that frequent them.

But the customer insight problem is two-fold, adds Gates. “The first part is knowing how to locate and access the right information to get customer insight. The second part is harder still: it’s about knowing what to do with it.” So to a large extent, real customer insight is a matter of a retailer figuring out what questions it needs to ask to take the business forward, based on careful analysis. “The modern retailer is primarily an information business,” he says. “Figuring out how best to exploit that information isn’t about tools or technology, it’s about people and questions.”

The good news for retailers is that the answers you need could be found somewhere within your own business – it’s simply a case of knowing where to look.