Anyone who has ever shopped on an e-commerce website will probably be familiar with recommender engines. These are the pieces of software that use your shopping history and other data to recommend other items you may be interested in. They were fairly rudimentary at the outset, recommending thousands of detective fiction titles to customers who had purchased one Ian Rankin novel, for example.
Time and technology have moved on and the use of advanced analytics, machine learning, AI and a good measure of behavioural science has made the systems for more sophisticated and productive. Today, retailers can predict quite accurately the likelihood of individual customers making a purchase and what they will be most interested in buying.
They are also using AI to create highly individualised marketing campaigns, develop new products and services targeted at particular consumer segments, to optimise the shopping experience, and bring new efficiencies to delivery services.
“With the rise of AI capabilities, there has been an increase in investment from organisations to integrate AI tools to deliver operational efficiencies and enhance consumer experiences,” says Richard Hepworth, EY Ireland consulting partner. “In the retail space, we are seeing a variety of applications across the end-to-end supply chain, from enhanced product development, optimised distribution and pricing, personalised marketing and promotions, focused employee training, efficient customer service and immersive shopping experiences.”
Making products that are more relevant, more efficient and run smoother with less energy consumption
How your KitKat break is helping cocoa-farming families close the living income gap through an innovative programme
Renewable gas will power Ireland’s decarbonised future
All aboard: Ten great things to see and do in Heritage Week
Some brands are investigating the benefits of AI across their operations, he notes. “Retail units and department stores are exploring AI to enhance the in-store experience. For example, they could use AI to analyse foot traffic and customer interactions to optimise store layouts and product placements, ensuring that high-demand items are easily accessible. Service providers have been investing in AI to improve their delivery services, which directly impact retail operations. AI helps in optimising delivery routes and predicting the best times for deliveries, which is crucial for online retailers to ensure customer satisfaction.”
Retailers do not even need access to personal data to enhance the consumer experience and increase the likelihood of a sale, according to Three Ireland head of digital Paul Prior. He explains that recommender engines look at the behaviour of the individual and at similar behaviours by others. The question is if enough people behave like that to offer useful insights into an individual’s future behaviours.
“A retailer will already have some information on that person from visits to their website and so on,” he points out. “They now have access to so much more information. There is data on propensity to purchase and so on. All that generic data is available commercially and we can utilise that to train AI to predict behaviour without having personal information.”
Data is collected from various sources, according to Hepworth. “They include customer data such as purchase history, browsing behaviour, and preferences; inventory data like stock levels, product movement, and demand patterns; transaction data including payment methods, transaction amounts, and frequency; and behavioural data from interactions with websites, apps, and in-store sensors.”
Technology is just part of the story. “It’s all about psychology and neuroscience,” Prior explains. “For example, we love talking to people because we have a need to interact with other people. In the digital world, when chatbots came out first they tried to make them sound like humans. But people were uncomfortable with that.”
At an individual level, the technology can enhance the prospects of a successful sale by monitoring the consumer’s activity on a website, he notes. It can look at the length of stay, identify if they are having trouble finding what they are looking for, and predict if they are likely to go elsewhere to look for it.
“If you can identify that you can try to intervene,” Prior adds. “For example, if they are looking at a new phone, it might be the camera features they are looking at. They might be a photographer, and we can make suggestions that will help them find what they are looking for.”
This is particularly important in light of other changes in consumer behaviour, he points out. “People have a much shorter attention span now and spend less time reading through content searching for what they are looking for. Generative AI is able to create highly relevant content very quickly and deliver it to customers. A lot of work goes on at the back end in terms of delivering that faster.”
Of course, people will have privacy concerns. “With regulators also taking steps to impose stricter controls on AI usage, for example through the EU AI Act which recently came into force in Ireland, and with clear GDPR regulations already in place to ensure data privacy, security and controls, many organisations have strong governance in place around the AI experiences they offer and the consumer data they use,” says Hepworth.
Consumers can take steps to protect their personal information, he adds. “They should review organisations’ privacy policies to understand how their data is being used and shared, utilise strong and unique passwords to protect their information stored on online accounts, and regularly monitor accounts for any unauthorised activity. Finally, consent remains a core principle of GDPR and consumers should take advantage of opt-in/opt-out options for data collection and marketing communications and ensure they are in control of what they are sharing to enhance their experience with the brands they trust.”
Ultimately, it’s all about making things better for customers, says Prior. “The technology enables to only offer customers products that are relevant to them and give them the experience they expect to have.”
And that means a highly individualised experience. “The type of journey most customers want tends to be very different than the one I would, for example. The best thing I’ve learned is that I’m not representative of anyone else.”