Success Metrics for Choice Overload in the Uber Eats Ordering Flow

You are a product manager on the Uber Eats food delivery app. Customers are experiencing decision fatigue: the app presents so many restaurants and dishes that they become confused and struggle to settle on an order. Which success metrics would you use to check its impact? For this prompt, decision fatigue means the confusion that comes from having a very large number of options to choose from. ```hint Where indecision leaves a trace A confused customer rarely triggers an error. Picture the event log of a session where someone cannot make up their mind, compare it with a session where they order quickly, and ask which differences you could count. ``` ### Clarifying Questions - Does "impact" mean sizing what decision fatigue costs Uber Eats today, judging whether a fix reduces it, or both? - Is the problem concentrated in dense markets where a very large number of restaurants deliver to the same address, or reported everywhere? - Which surfaces are in scope: the home feed, search results, long restaurant menus, or the whole ordering flow? - Which behaviors does the app already log, such as store page views, scroll depth, cart edits and abandoned carts? ### What a Strong Answer Covers - A behavioral definition of a fatigued session, and a way to tell it apart from look-alikes such as finding nothing appealing, fee shock at checkout, or browsing with no intent to order. - One primary success metric tied to whether ordering intent turns into a placed order, supported by diagnostic metrics along the browse-to-checkout funnel. - Guardrails that catch a fix that narrows choice too far: basket size, order frequency, order ratings and how evenly orders spread across restaurants. - A plan to size the orders lost today and to attribute changes to a fix through segmentation by choice density and a controlled experiment. ### Follow-up Questions - After you shorten the home feed, time to order falls but average basket value also falls. Is that a success? - How would you separate decision fatigue from customers who open the app just to compare prices and delivery times? - The fix lifts conversion for new customers, but tenured customers try new restaurants less often. What would you do?