Google Cloud debuts new retail AI tools ahead of NRF 2023

NRF 2023, the retail industry’s largest event — presented by the National Retail Federation — opens on Monday at the Javits Convention Center in New York City. But in advance of what is known as “Retail’s Big Show,” today Google Cloud introduced a variety of new and updated artificial intelligence (AI) technologies to help retailers boost in-store inventory shelf-checking, enhance online shopping, provide more personalized search and offer better recommendations.

According to Amy Eschliman, managing director of retail solutions at Google Cloud, since the pandemic, shoppers want a more fluid and natural shopping experience online.

“Before the pandemic, 80% of transactions happening globally were in-store, but the shift to digital was continuous; COVID flipped the switch overnight,” she told VentureBeat by email. “While in-store shopping has definitely resumed, the shopper is forever changed.”

Making online shopping more personalized and intuitive

To meet the new consumer expectations, Eschliman explained that the new AI-driven personalization capability customizes the results a customer gets when they search and browse a retailer’s website.

It uses a customer’s behavior on an ecommerce site, such as their clicks, cart, purchases and other information, to determine shopper taste and preferences. The AI then moves products that match those preferences up in search and browse rankings for personalized and relevant results. 

“We know shoppers want this sort of personalized experience more than ever,” she explained. She added that research commissioned by Google Cloud found that 75% of shoppers prefer brands that personalize interactions and reach out to them, and 86% want a brand that understands their interests and preferences. 

Transforming digital window shopping

Browse AI is Google Cloud’s new feature in its Discovery AI solutions for retailers, which uses machine learning to select the optimal ordering of products on a retailer’s ecommerce site once shoppers choose a category, like “women’s jackets” or “kitchenware.”

Historically, ecommerce sites have sorted product results based on either category bestseller lists or human-written rules, like manually determining what clothing to highlight based on seasonality.

Browse AI takes a new approach by self-curating and learning from experience, saving retailers the time and expense of manually curating multiple ecommerce pages.

The new tool is now generally available to retailers worldwide, supporting 72 languages.

Google Cloud’s AI-powered shelf checking

According to a NielsenIQ analysis of on-shelf availability, empty shelves cost U.S. retailers $82 billion in missed sales in 2021 alone.

Built on Google Cloud’s Vertex AI Vision and powered by two machine learning models — a product recognizer and tag recognizer — Google Cloud’s new AI-powered shelf-checking solution is available globally in preview. And it helps solve a thorny problem: how to identify products of all types, at scale, based solely on the visual and text features of a product, and then translate that data into actionable insights.

Eschliman explained that the solution utilizes Google’s vast database of facts to give retailers the ability to recognize billions of products, to ensure their shelves are stocked properly. “This comprehensive dataset, paired with Google Cloud’s state-of-the-art AI, can help retailers better manage their in-store inventory,” she said. And it tackles “the legacy industry challenge of retailers knowing what their shelves actually look like at any given time, and where restocks are needed.”

AI ups the retail recommendation game

Google Cloud also added upgrades to Recommendations AI, announced today, to make ecommerce even more personalized and dynamic.

A new page-level optimization feature now enables an ecommerce site to dynamically decide what product recommendation panels to uniquely show to a shopper. Page-level optimization also minimizes the need for resource-intensive user experience testing, and can improve user engagement and conversion rates.

In addition, a recently-added revenue optimization feature uses a machine learning model, built in collaboration with DeepMind, that combines an ecommerce site’s product categories, item prices, and customer clicks and conversions to find the right balance between long-term satisfaction for shoppers and revenue lift for retailers.

Finally, a new buy-it-again model leverages a customer’s shopping history to provide personalized recommendations for potential repeat purchases.

Retailers can get buried in data

Many retailers are still early in the process of truly leveraging their customer, product and supply chain data in real time to improve business operations and customer experience, said Eschliman.

“But the reality is that it’s easy to be buried in data in retail,” she said. “AI and machine learning are uniquely qualified to tackle the challenges retailers face today because the technology is able to process and analyze large amounts of data in real time, identify patterns and trends, and make predictions and decisions with an increasingly higher degree of accuracy and reliability.”