Root-Cause Analysis of High Rice Returns at a Single-City Grocery Delivery Startup

You are the product manager at a hyperlocal grocery delivery startup. Customers send back rice far more often than the business expects: they order it, the order is delivered, and then the rice is returned. The facts you have: - Rice returns have been high ever since the startup launched; this is not a recent change. - The startup operates in only one city, and the problem shows up across that whole city. - Rice is the only affected category. Other groceries are not being returned at an unusual rate. Explain what might be causing the returns, how you would find the actual root cause, and how you would solve it. ### Clarifying Questions - What counts as a return: a customer refusing the bag at the door, sending it back after accepting it, or both? Can a bag that has been opened be returned? - High compared with what: the return rate of similar staples such as flour, lentils or cooking oil, or a target the startup set for itself? - Is rice picked from the startup's own small warehouses or bought from partner neighborhood shops, and is it sold only in sealed branded packs or also loose or repacked? - Does the return flow record a reason, and is the refund issued before anyone checks the returned item? ### Part 1 — Generate the hypotheses List the plausible root causes of the rice returns. Organize them so the list is complete, then use the three known facts to say which causes are most and least likely. ```hint Let the facts prune the list A cause that began with some later event cannot explain returns that date from launch, and a cause that touches every grocery item cannot explain why only rice comes back. Ask what sets rice apart, both as a product and in how it is listed, handled and used. ``` #### What This Part Should Cover - A structured set of causes that follows the customer from choosing rice in the app to cooking it. - Explicit use of the launch-day onset, the single city and the rice-only pattern to rank causes and set some aside. - Causes specific to rice rather than generic grocery problems, including incentive and policy effects. ### Part 2 — Find the root cause Describe how you would test your leading hypotheses: which data you would pull, which checks you would run outside the data, in what order, and what result would confirm or rule out each cause. ```hint The returns carry evidence Think about what the state of a returned bag, and the time between delivery and return, would each tell you, and about which grocery categories make a fair comparison for rice. ``` #### What This Part Should Cover - Confirming that the return metric is defined and logged correctly before diagnosing anything. - Cuts and comparisons that separate competing causes, with a baseline from comparable staples. - Qualitative sources and a sequence that tests cheap, likely causes first. ### Part 3 — Solve it Propose fixes for the causes you expect to confirm, decide which to ship first, and explain how you would know they worked. ```hint No second city to compare Because the startup runs in one city, there is no untouched market to act as a control. Think about what else could serve as a baseline when you measure a fix. ``` #### What This Part Should Cover - Fixes matched to each confirmed cause rather than one blanket fix. - Prioritization by impact, cost and speed, including the trade-off between stricter returns and customer trust. - A success metric with guardrails and a way to measure impact within a single city. ### What a Strong Answer Covers - One hypothesis-driven thread from causes to tests to fixes, rather than three disconnected lists. - Disciplined use of the scenario's facts: onset at launch, a single city, rice alone. - Grounding in how households buy and use rice: many look-alike varieties, several pack sizes, heavy bags, quality that shows only after cooking. - Attention to how the return policy and promotions shape customer behavior, alongside product and operations causes. - Clear success and guardrail metrics. ### Follow-up Questions - Your data shows that most returned bags come back sealed within a few hours of delivery. What does that suggest, and what would you check next? - Finance proposes refusing all rice returns from now on. How do you respond? - How would you keep the same problem from appearing when the startup adds a second city or a new staple category?