As you most likely are aware, Instacart provides shopping and home delivery in a variety of stores.
In an ongoing effort to generate more precise shopping trips — i.e., a faster way of shopping for its employees, who are shopping on behalf of customers — the San Francisco-based company has been testing various ways to determine how people might most efficiently shop for items on a list, ranging from:
- an alphabetical list
- a route-based approach
- an artificial intelligence (AI) approach that uses data from the company’s most efficient shoppers to predict a sequence of picks that would be the most efficient
The article states that people are using AI to solve hard problems more and more… and the algorithms used for more traditional problem solving are not so different from those that can determine how a human would pick-up specified items in a store. In fact, Instacart is able to “guess” the next item a shopper will pick 60% of the time with the AI solution!
“It’s not 95%,” said Jeremy Stanley, VP of data science at Instacart. “But there’s room for variance and error. When we look at the overall sequence it mimics what the shopper does very closely, and usually only reverses a few items per trip.”
Might there be an application for using this AI approach for you and me to enhance our food shopping experience?
If so, how might retailers view this developing use of technology, considering it might make shoppers less susceptible to impulse buying because they will do less wandering through the aisles?