Uber Ride Ratings Trail the Industry by 20%: Diagnose the Gap and Close It

You are a product manager at Uber. Uber's ride rating is running 20% below the industry standard. How would you work out what is behind the gap, and what would you do to close it? Walk through how you would confirm the gap is real, find its causes, choose which fixes to pursue first and measure whether they worked. ```hint Check the yardstick first Before hunting for causes, pin down exactly what is being compared with what. Then follow a trip from request to drop-off and look for the stage where low-rated rides differ most from well-rated ones. ``` ### Clarifying Questions - What does "ride rating" mean here: the stars a rider gives a trip once it ends, the app-store rating, or a survey score? Is the 20% gap in the average score or in the share of top ratings? - Where does the industry standard come from: which competitors and which markets, and do they ask for ratings with the same scale, timing and defaults as Uber? - Has the gap existed for a long time or did it open recently, and is it concentrated in particular cities, ride products such as shared or premium rides, or app platforms? - Can riders attach reasons, tags or comments to a low rating today? ### What a Strong Answer Covers - Validation of the gap before any fix: the exact metric, how comparable the benchmark is, and whether rating-flow design or the kind of rider who chooses to rate explains part of it. - A structured breakdown of possible causes along the trip (booking, waiting and pickup, the ride, drop-off and payment) plus external factors, each paired with the data that would confirm or rule it out. - Segmentation of low ratings by market, ride product, driver, vehicle and trip attributes to find where the gap is concentrated. - Fixes tied to confirmed causes on both the rider and driver sides of the marketplace, prioritized by impact, confidence and effort. - A success metric for the gap, with guardrails such as driver supply, wait times, trip volume and rating response rate. ### Follow-up Questions - Suppose most of the gap comes from how Uber asks for ratings rather than from worse trips. Would you change the rating flow, and how would you explain the 20% to leadership? - Removing the lowest-rated drivers would raise the average quickly. What could that do to supply, wait times and fairness, and how would you decide? - Riders who leave ratings are a self-selected group. How would you check whether ratings reflect the experience of riders who never rate?