James David Williams was a pioneer of what is now known as multi-channel retail. In 1859, he set up three ‘mobile shops’ in Manchester, selling assorted goods from wheelbarrows. When the Post Office started carrying parcels in 1882, he spotted an opportunity and the business went on to become one of the largest catalogue retailers in the UK.

In the past two decades, the company – now trading as JD Williams and owned by N Brown Group Plc – has built an online presence that today accounts for around half of its sales. It has accumulated numerous clothing brands, including High & Mighty, Simply Be and Figleaves.

And now, in contrast to the traditional retail move from bricks and mortar to online, the company is preparing to opensome of its first-ever physical stores. Every new channel represents a new source of customer data, which provides countless opportunities to offer personalised offers and services across channels, but only if the data can be integrated and analysed as one.

Online Analysis

Most recently, JD Williams undertook a project to derive greater insight from its web traffic data. “When I joined three years ago,” explains IT director Neil McGowan, “one of my main challenges was to get a better understanding of what our customers do online.”

The company already collected extensive aggregate traffic data – it attracts a million customers a week across all its sites, generating 21 million page views. But at the time it could not drill down into an individual visitor’s journey through the site. This prevented JD Williams from answering some important questions. For example, why do people add items to their online shopping cart but then leave the site without paying?

It was the desire to answer this question that McGowan used as the business case for investment in clickstream analysis technology. Whereas onventional web analytics technology counts page impressions, clickstream tools collect individual users’ navigation path, generating particularly large quantities of data.

McGowan identified Celebrus as a potential supplier, in part because its technology integrates easily with the Teradata customer data model that JDWilliams uses, and ran a pilot project to evaluate the product.

Even after the pilot, JD Williams was able to understand why potential customers abandoned their shopping carts before checking out. “We could see whether individual customers were coming back and completing their purchases, either online or through other channels such as the call centre,” explains McGowan. “And we were able to do marketing campaigns based on people who hadn’t responded.”

The improvements in sales that resulted from this analysis were enough to justify the investment alone.
Since then, the company has conducted further analyses, including an investigation into how many customers are using mobile devices to visit its sites (2.8% on average, including 0.8% through iPads). All this information, McGowan says, comes as a bonus on top of the original business case.

Opening up shop

McGowan is now considering the impact that the new stores will have on the company’s IT systems. The first stores, based on the Simply Be womenswear brand, will be ‘flagship’ premises in the UK’s major cities, and are designed as much to promote the brand as to shift merchandise.

Multi-channel integration will allow customers to be measured in store, and then buy clothes online that they know are going to fit them. “The biggest inhibitor to buying clothes online is not knowing if they will fit,” explains McGowan.

Quite how customers will authenticate their accounts has yet to be decided, McGowan explains, but there is no need for an extra repository of customer information to support the physical channel – the infrastructure that JD Williams has built for its online and catalogue businesses can accommodate one more channel.

Indeed, he says, the ideal is for there not to be multiple channels at all, but instead one highly flexible channel – a so-called ‘omni- channel’ approach. “Our challenge is to manage the business in one channel, and prevent silos from splitting off,” he explains.