Make Retail Personal with AI-Powered Product Recommendations

Why Consumers Want More Personalization

According to Salesforce, one-third of consumers report that being offered products that aren’t relevant to them is a top frustration. That makes sense: Consumers want personalization, and they’ll go, or stay, where they can get it. In fact, two-thirds of consumers say they’ll remain loyal to a company if it offers a personalized experience.

DISCOVER: The four most effective uses of generative AI in retail.

To date, however, most retailers haven’t capitalized on this. In a survey of more than 100 brands and retailers, only 20 percent report customizing product recommendations based on a customer’s purchase history. That poses a problem. According to McKinsey, consumers don’t just want personalization, they expect it: 67 percent cite relevant product recommendations as an important personalization action when making a first-time purchase. A lack of recommendations can foster a lack of loyalty, with 3 in 4 consumers getting frustrated when they don’t find personalization from a brand.

In fairness, providing personalized recommendations isn’t easy, particularly because it has to fluctuate. People’s tastes change — and they don’t just want the familiar; they want something new too. But with 65 percent of customers expecting companies to adapt to their changing needs and preferences, they want a product that is uniquely tailored to them.

In a sense, consumers want retailers to give them the same personalized experience they receive from Spotify. The music platform uses AI for personalization and recently turned to Google’s artificial intelligence to tailor podcast and audiobook recommendations. Similarly, retailers may find generative AI to be the solution to providing personalized recommendations that meet customer expectations.

How AI Drives Better Product Recommendations

It’s one thing to collect customer data and identify patterns in real time. Using these insights to generate real-time personalized product recommendations is another, harder thing. That helps explain why, as Salesforce notes, “only 32% of retail executives say they can turn profile information, purchase history, and service interactions into tailored experiences that make shoppers feel like VIPs.” Fortunately, generative AI is making it easier.

By generating new data points from learned patterns, generative AI can take the guesswork out of product recommendations, serving up items that align with individual preferences. Just as ChatGPT analyzes inputs and sifts through its data bank to offer near-instant responses to individual queries, generative AI in retail makes it possible to go beyond merely recommending the top sellers and instead tailor recommendations to each shopper. So, for example, while a customer may add a ceramic mug to his cart, AI can promptly assess whether he drinks coffee, tea or water and make recommendations accordingly.

FIND OUT: How AI and machine learning solutions can help improve data management.

Mastercard recently launched Shopping Muse, a generative AI tool that translates colloquial language (e.g., “beach formal”) into tailored product recommendations based on a user’s profile. And with digital signage, such tools may cross over into physical stores sooner than later. Broadly speaking, however, AI-powered product recommendations aren’t in the distant future.

For example, 35 percent of purchases on Amazon in recent years were the result of the retailer’s product recommendations, powered by generative AI. And during the Prime Big Deal Days last fall, Amazon’s AI-enabled product recommendations played a significant role in boosting sales across the health and beauty categories.

AI-powered product recommendations are already here, and their widespread use is on the horizon. The retail industry is on the road there.