Data in Action: Predicting Customer Order Probabilities
Founded in 2014, Fresh Prep, a Vancouver-based meal kit delivery company, is all about helping their customers cook quick and healthy dinners. Fresh Prep offers a rotating menu of 10 different recipes every week.
Working with a group of UBC Master of Data Science (MDS) Vancouver students for their capstone project, Fresh Prep initially wanted to get a prediction of how many orders they would be delivering the next week. However, the MDS team dug even deeper into the data and were able to also predict which customers were likely to order in the weeks ahead.
The students developed a dashboard that will allow Fresh Prep to easily manipulate the data and see insights from multiple angles. The dashboard consists of a descriptive panel that shows past orders and a predictive panel that shows future orders. The dashboard can also be filtered by customer type, date, dietary restrictions, subscription type, and location.
With this information on hand, Fresh Prep will be able to delve deeper into potential strategies for marketing, production, and business, and improve their order rate by targeting specific clients who are unpredictable and potentially getting them to order consistently. It will also allow the company to avoid sending excess emails to their existing clients.
While this project only touched on a very small piece of the total data Fresh Prep had, it has the potential to be expanded in the future to include additional questions, like whether or not a specific recipe could have an effect on a customer’s decision to order again.