Dynamic Yield, which uses artificial intelligence to power personalization, on Wednesday announced the launch of an Identified Offline Purchase Data Ingestion feature that allows brands to import lists of offline purchases that take place in physical stores.

The technology, available in the company’s Data Loader feature, matches the records with online identities to enable brands to run personalized campaigns based on offline purchases. It brings offline data online to improve targeting and the quality of recommendations, providing consumers with a shopping experience that reflects their activity in stores and online.

“The offline data request was one of the top requested features,” said Liad Agmon, CEO of Dynamic Yield. He calls the strategy “omnichannel commerce,” which integrates offline purchases to provide interesting data points that don’t exist online. “Some of our smarter customers asked us to use the offline data and signals to create better online experiences.”

For example, the shopping pattern for a brand in a Philadelphia store is different from a store in Los Angeles. The platform uses the data, so when a customer arrives from the Philadelphia area to shop in a Los Angeles store they would receive different recommendations than if they were shopping in Philadelphia.

Since McDonald’s acquired Dynamic Yield in early 2019, the company added about 100 retail brands to its customer list for a total of 300, Agmon said. About 270 people at the company now define the company’s product roadmap and direction. 

And while there is a product roadmap for the company, the direction is a bit different for McDonald’s compared with other types of businesses such as retail. “We keep separate roadmaps for the restaurant business and retail,” he said.

There is a list of about 200 capabilities and ideas on a wish list that Agmon said his group must prioritize.

In this release, the platform can exclude recommendations for products already purchased in the store from online recommendations to eliminate marketing redundancy and increase relevancy during the shopping experience.

The new feature also can augment the data used in recommendation algorithms by using in-store data of items that are often purchased together. And it can combine offline purchase data with a user’s online shopping behavior to create a rich profile of their preferences and interests for more granular targeting.