When placing an Online Grocery order, the customer journey typically begins with choosing a date and time slot. For Express, we present the option for faster delivery in real time while the customer is going through checkout. It’s a simple option that customers can easily see, but there’s a ton of technology working behind the scenes in order to determine if that option should be presented to the customer.
Using machine learning and sophisticated algorithms developed by our Global Technology team, we’re able to consider thousands of variables, such as the number of items in an order, staffing, the types of delivery vehicles available and estimated drive times between our store and a customer’s address. We even pull in data for current weather conditions when estimating just how quickly we can deliver the order. The timing of how quickly orders are first prepared and then delivered become learning signals that feed back into our algorithms, so we can further improve our timing estimates for future orders.
Rolling out this new feature required coordination and agility in order to build a solution as fast as possible. By empowering a core team of people from tech, business, product and design, we successfully tested, released and scaled this capability in just one agile sprint over two weeks. As we continue to add new machine learning-driven capabilities like this in the future, as well as the corresponding customer experiences, we’ll be able to iterate and scale quickly by leveraging the flexible technology platforms we’ve developed.