Market Sizing: Swiggy's Nationwide Delivery Throughput in a Single Hour

You are asked an estimation question in a product manager interview: about how many orders does Swiggy, the Indian on-demand delivery app, deliver across India in one hour? Walk through how you would structure the estimate, the assumptions you would make and the arithmetic, and explain how you would check that your answer is reasonable. ```hint Build up from people Instead of trying to recall a company figure, start from the people in India who order delivery and how often they do it, reach a daily total, and only then convert it to hours. Ask yourself whether every hour of the day really looks the same. ``` ### Clarifying Questions - Should the estimate count only restaurant food orders, or also Swiggy's grocery and other delivery services? - Do you want an average hour on a typical day, or a busy hour such as weekday dinner time? - Should cancelled or failed orders be left out, so that only completed deliveries count? ### What a Strong Answer Covers - Scoping before estimating: which services are included, delivered versus placed orders, and average versus peak hour. - A clear equation that chains population to ordering customers, order frequency, Swiggy's share and the completion rate. - Assumptions labeled as assumptions, with round numbers, segmentation where ordering behavior differs, and careful arithmetic. - Converting a daily figure into hourly volume using a realistic time-of-day pattern. - A sanity check from an independent angle, such as restaurants or delivery partners, plus a note on which assumption the answer is most sensitive to. ### Follow-up Questions - How many delivery partners would need to be online during the dinner peak to handle your estimate? - How would the number change on a rainy evening or during a major cricket match, and what would that mean for operations? - Which single assumption is your estimate most sensitive to, and what data would you use to tighten it?