Google Cloud brings its search smarts to retail customers

Google Cloud has released a new fully managed search tool, called Retail Search, which is aimed at retailers who are concerned about product discovery on their digital platforms.

The service is intended to solve a common retail challenge of losing customers when they don’t immediately find their desired product on an ecommerce platform. This phenomenon, which Google calls search abandonment, could cost US retailers around $300 billion each year, according to research conducted jointly by Google and The Harris Poll. That research also found that 94% of US consumers had abandoned a shopping session because they received irrelevant search results.

Bringing Google Search to Retail Search

The new service, which was initially announced last year, aims to bring Google’s expertise around search indexing, retrieval, and ranking to a retailer’s unique product catalog and user data.

“It’s a full-stack service which has all capabilities of what a search engine does, rather than individual components of just ranking or just indexing or just scoring and so on,” said Srikanth Belwadi, group product manager at Google Cloud. “There is no need to pre-process data, train or hypertune machine learning models, nor load balance or manually provision your infrastructure to handle unpredictable traffic spikes. We do it all for you automatically.”

Other features include semantic search to effectively match product attributes with website content for fast, relevant product discovery, and optimization of product results based on user interaction and ranking models.

Retail Search builds on Google Cloud’s existing retail-focused solutions, such as Recommendations AI and Vision API Product Search. Released in July 2020, Recommendations AI is another managed service that aims to improve product recommendations. The Vision API computer vision product was launched in 2019 and provides retailers with the option of giving customers the freedom to search for a product using an image.

Retail Search: a market share and data play

According to analysts, Retail Search is aimed squarely at retailers trying to pivot digitally and offer competitive online shopping services.

“For retail specifically, it can help businesses make sure that customers quickly find the product or service they’re looking for, or that a similar product can be accurately recommended, to increase conversion, cross-sell, and upsell,” said Hayley Sutherland, senior research analyst at IDC.

For Forrester’s senior analyst Tracy Woo, Google’s intent is to gain more market share while also gaining access to more data. “These are machine learning algorithms; the more data collected the better the results,” she said. “Embedding their search engine in retailers is a two-way street. Retailers benefit from Google’s superior search capabilities; Google gets all of the data gained from these search results.”

Google reached out after publication to clarify that its Retail Search models are specific to a customer’s data, which is strongly isolated and not used to improve Google Search services.

Ray Wang, principal analyst at Constellation Research, sees Retail Search as a strongly differentiated product to anything offered by rival cloud vendors Amazon Web Services and Microsoft Azure.

“Google Cloud is empowering customers with tools to compete against Amazon in an effective manner. As you know, most retailers will not use AWS for cloud and Google is capitalizing on this with a cross-sell. Customers get Amazon.com capabilities, without having to go to Amazon and AWS,” he said.

Retail Search is now generally available to Google Cloud customers.