How machine learning, AI improve the luxury retail experience

Artificial intelligence (AI) and machine learning are being introduced and utilized by nearly every industry. This includes the luxury retail sector, where advanced algorithms are allowing retailers to enhance the customer experience both in-store and online.

Brands such as Dior have launched AI chatbots that can interact with and make recommendations to online shoppers, while Burberry’s flagship store in London has fully embraced the digital age with interactive multimedia tools such as mirrors that double as screens.

In-store, luxury retailers are using AI and augmented reality (AR) to provide customers with a more personalized and immersive shopping experience. An example of this is smart mirrors, which can automatically recommend accessories for given outfit, as well as allow shoppers to make a 360-degree video and change the color of clothes.

While luxury retailers still rely on human sales associates for the time being, AI and machine learning are playing an increasingly larger role in helping brands interact with their customers.

As a report by McKinsey & Company states: “Big data and machine learning are bringing authenticity and relevance back into the customer relationship,”by providing advanced analytics to help brands offer services uniquely tailored to each customer and occasion.

According to Andrew Fowkes, Head of Retail Centre of Excellence at SAS UK and Ireland, “applying machine learning to in-store data has great potential to help luxury retailers better understand their customers through their entire lifecycle. He added retailers were also “utilizing these techniques to better understand demand patterns, to have the high-value merchandise in the right part of the world to fulfill demand.”

Fowkes said luxury retailers were deploying machine learning techniques to better understand their customers and keep them engaged with a brand. Such techniques were also used to “apply more real-time context to offers or messages they put in front of their customers, or the messages their employees can use to sell more when face-to-face with the customers,” he added.

Ray Hartjen, Marketing Director at RetailNext, emphasized the need for accurate footfall data, noting the importance of deep-learning based sensors such as RetailNext’s Aurora v2, which can accurately distinguish between customers and reflections, shadows and shopping carts piled high with merchandise. Hartjen added that advanced sensors and processes were able to determine shoppers from sales associates, providing not only accurate footfall, but also information on how, when and where shoppers and staff interact.

“Inside the store, sensors with AI determine what shoppers are doing at displays other than dwelling within a certain geo-fenced location. For example, the sensor can determine if a shopper reaches for an item, picks it up, looks at it more closely or tries it on, returns it to the display, etc. Deep-learning based human activity recognition delivers the data that allows for retailers to modify their store layouts, displays, fixtures, product assortment, staffing models — everything really — to drive the desired outcomes they’re designing toward,” he said.

Fowkes also pointed to the use of computer vision; a new discipline that trains machines to interpret and understand the visual world using digital images from cameras combined with deep learning models that mimic the processes used by the human brain.

“Our most developed customers can join online browsing data, social media influence and even images deploying computer vision techniques to automatically generate attributes. These attributes can then be used to fine tune customer real-time offers or future design and development of products,” Fowkes said.