Online grocer applies machine learning to shopping lists
Farmstead is turning shopping lists into a highly targeted promotional tool with predictive analytics.
The San Francisco-based online grocer is introducing a new feature called Smart Shopping List that goes beyond recommendations to predict what customers need. Smart Shopping List bases predictions on multiple data-driven factors, including weekly shopping history, buying signals, and what’s already in their cart.
Smart Shopping List dynamically generates an endless list of heavily personalized products that customers can scroll through. These products are ranked by machine-learning algorithms based on customers’ shopping behavior and wellness preferences, along with the collective behavior of other customers who buy similar products. Farmstead intends to enable customers to put their orders together faster with assistance from its predictive models.
Farmstead’s service also algorithmically considers more complex factors such as day of the week and seasonality (such as heavy cream during Thanksgiving) and compatible products (such as cream cheese a customer normally buys when they buy bagels), and programmatically surfaces seasonal products (such as fresh local strawberries in the summer).
Through machine learning, Smart Shopping List becomes more personalized every time a customer uses it. In early testing, Farmstead found that more than 40% of customers choose to shop via Smart Shopping List when presented with the option, and when they did they were 3.5 times more likely to add suggested items to their list.
Smart Shopping List is available now to existing Farmstead customers. New customers will be presented with the Smart Shopping List option after they have placed their first order with Farmstead, giving the software the chance to learn their preferences.
“One of the biggest challenges to buying groceries online is to build your shopping cart rapidly,” said Farmstead co-founder and CEO Pradeep Elankumaran. “Since there are no physical aisles to walk through like a traditional supermarket, and only so much scrolling one can take, it’s a true challenge to help customers find the right products at the right time while also suggesting new ones they’ll love in the few minutes they spend each week tapping buttons online.”
Farmstead also recently announced a predictive analytics-based program called Refill & Save that lets customers set up weekly subscriptions for their staples. Refill & Save provides customers with recurring orders with discounts on certain staples, including milk, eggs, fresh fruit, fresh vegetables and bread. To create savings, Farmstead uses artificial intelligence-powered predictive models to improve efficiency and reduce costs in its supply chain.