New dunnhumby tech uses AI for product selection
Planograms, localized assortment, and predictive modeling featured in new suite of inventory tools
![assortment_blog_image1.jpg assortment_blog_image1.jpg](https://eu-images.contentstack.com/v3/assets/blt58a1f8f560a1ab0e/blt36fc3959f8759c92/66cc876265cbc37e5d9ebba2/assortment_blog_image1.jpg?width=1280&auto=webp&quality=95&format=jpg&disable=upscale)
Customer data analytics firm dunnhumby has launched a new suite of tools that use artificial intelligence to localize assortment selection and optimize planograms, the company announced on Monday.
“In today’s dynamic retail landscape, keeping up with customer expectations while outpacing local competition is crucial,” said Jenn Dabbelt, dunnhumby global head of product, in the press release. “Our new assortment tool, powered by advanced AI, allows retailers to build store-specific assortments centered on what the customer wants.”
The AI model uses purchase data to determine which products are best for individual stores. The company noted that some products sell better than others, noting that protein shakes might sell better at stores near a gym, for example.
“By looking at the relationships between products and stores, advanced recommender systems can detect relationships and highlight opportunities with a much lower risk of error,” the company said.
The planogram tool takes a similar analytical approach at the shelf level, using historical data to “choose assortments that optimize sales potential …”
The new tools also feature predictive AI that estimate what would happen when products are introduced or removed from a store.
“For example, removing a certain pack size or flavor of cookie will not necessarily reduce category revenue by its full sales value. dunnhumby’s AI lets retailers understand how much of those sales would transfer to a similar remaining product in the assortment, creating a more accurate forecast,” the company said.