Supermarkets

Morrisons launches AI tool to help shoppers find products

Peter Laflin, director of data at Morrisons, told City Am the system was designed and hosted internally to improve security and scalability

Register to get 1 more free article

Reveal the article below by registering for our email newsletter.

No spam Unsubscribe anytime

Want unlimited access? View Plans

Already have an account? Sign in

Morrisons has reportedly launched a generative AI search tool to help customers locate products in store, according to City AM

Developed in-house using Google Cloud’s Vertex AI platform and Gemini large language models (LLMs), the feature already handles more than 50,000 daily searches during peak periods. This is said to allow shoppers to type queries such as “tahini” or “that tomato puree in a tube” and receive real-time aisle-level product locations. 

Peter Laflin, director of data at Morrisons, told City Am the system was designed and hosted internally to improve security and scalability.

The supermarket is among a number of UK and US grocers investing in AI tools to enhance operations and customer experience. Sainsbury’s is preparing to launch its in-house media platform Pollen, which uses AI to optimise campaigns with Nectar data, while Walmart has also invested heavily in similar technology.

According to Morrisons, its decision to develop the tool internally followed concerns about third-party vulnerabilities. Last year, the retailer was affected by a ransomware attack on its logistics partner Blue Yonder, and in June Marks and Spencer confirmed that data from some customer accounts had been compromised following a breach at a third-party provider.

Laflin described in-house hosting as “non-negotiable”, adding, “we can be confident that our data is secure, it’s always at the forefront of what we do.”

He also said the idea for the search feature arose from shoppers frequently asking staff where certain items can be found in the store. 

The system is trained to recognise typos, brand references and ambiguous phrases, and returns the closest matches based on each store’s layout and inventory. Laflin said the team focused on addressing everyday pain points for shoppers, particularly when looking for niche items that may not be well signposted. 

Check out our weekly podcast: 'Talking Shop by Retail Sector'

Back to top button
Secret Link