AI-enabled retail – helping to level the playing field for physical retailers

The rise of ecommerce and subsequent access to a whole lot of data about online shoppers has provided retailers with new insight into customers shopping habits and behaviours and as a result, has changed the way retailers engage with their customers, both online and in store. In the past, it has been difficult for retailers to apply digital data to the in-store environment which has meant that online retail was always a step ahead in terms of customer intelligence and personalisation. However, the tables are turning and modern bricks and mortar retailers are catching up. Customer data and the high street Modern store systems are now capable of aggregating massive data sets from nearly every consumer touchpoint – digital or physical – and so, have emerged as powerful competitive assets. AI-enabled applications, built on machine learning, can evaluate trillions (yes trillions!) of customer data combinations – far more than any human or traditional enterprise system ever could – to deliver precise recommendations to retailers, CPG manufacturers and customers alike. By removing the chance of human bias and the data limitations of the past, bricks and mortar retailers can better optimise their marketing strategy in real-time, based on accurate data. This brings about improvements to the shopper experience and further strengthens the relationship between brands and their customers. Anticipate what the consumer will want next The greatest benefit of AI-enabled applications, however, is that they can identify trends or anomalies among those trillions of data points to predict future behaviour faster and more accurately than a team of analysts. These applications are able to project outcomes and guide marketing decisions based on any combination of potential inputs. For instance, a targeted discount for a particular product or the introduction of a private label alternative to a low-margin supplier item. At the consumer level, rule-based algorithms are limited by their understanding and segmentation of historical data. Although marketing campaigns using this foundation are today considered ‘personalised,’ they are in fact not comprehensive and often result in a flat level of redemption and only incremental growth rates. In contrast, personalised marketing solutions that use AI can anticipate consumer needs by analysing trillions of combinations per household and identifying points in historical data that help to predict future behaviour. By giving the system the freedom to choose from an unrestricted range of offers that humans may not think to send, the system is able to ‘learn’ the best mix of offers for increasing redemption rates and supporting marketing goals. Making the right offer at the right time for the right person No matter how intelligent the targeting, marketers need to ensure their messages are reaching consumers at exactly the right time and in the right place. An AI-enabled system will work out that a customer is likely running low on pet food, for example, and will be able to suggest an alternative brand with healthier ingredients that will deliver higher margins in the long run. By sending a targeted offer through a mobile app, triggered by a geofence – or virtual geographic boundary – marketers can engage with shoppers in-store, where they are likely to respond much more positively than if they were just to receive a coupon in the post or at the point of sale. As you might imagine, a customer’s willingness to redeem an offer is not just about the depth of the discount; it’s just as much about the relationship the customer has with the product or brand in question. But that doesn’t mean the customer must be brand loyal or must even have purchased the product before. In fact, they don’t even need to have ever considered a purchase from the category. In a standard merchandising model, no one could predict when a customer who has never bought a sports drink, or any drink, would be willing to enter the category. But AI-enabled category management solutions – assigned the task of increasing sports-drink sales – can identify trends in behaviours for sports-drink consumers in other categories that make it clearer who to target and what it would take to convert them. A data-driven future for the high street Today, physical retailing is all about creating a store environment built on rewarding, data-driven customer engagement strategies that lead to long-term loyalty. No longer are personalised marketing or curated assortments the exclusive domain of digital, nor is it any longer acceptable to create offers that support business goals but which don’t fly with shoppers. There is now a fundamental mandate for stores to have a 360-degree understanding of their customer and to know precisely how to act upon that information. Fortunately, with the arrival of AI, synthesising and acting upon the insights customers share at every touchpoint becomes easy. Doing so will create a new relevance and strength for physical stores in a data-driven digital economy. By Julian Bridle, Vice President Presales, EMEA at Symphony RetailAI