Root-Causing Slower Instamart Deliveries in a Tier-1 City Since Last Week
SwiggyProduct Manager · OtherLast updated
A product manager metric-diagnosis question about Swiggy Instamart orders in a Tier-1 city taking 10% longer than the expected standard delivery time since last week, across every order type. It tests metric validation, breaking delivery into stages, segmentation, and separating internal from external causes.
You are a product manager at Swiggy working on Instamart, its quick-commerce grocery delivery service. Since last week, Instamart orders in a Tier-1 city have been taking 10% longer on average than the expected standard delivery time. The slowdown is not confined to one kind of order; it appears across every order type. Find the root cause of the increase.
Clarifying Questions
- Is the expected standard the delivery time promised to customers at checkout, or an internal benchmark, and has that standard or the model that sets it changed recently?
- Is this city the only one affected, or do other cities show a smaller version of the same rise?
- Did the increase start abruptly on a specific day last week, or build up gradually over the week?
- Is it spread evenly across the city's dark stores and hours of the day, or concentrated in certain stores, neighborhoods or peak hours?
Hints
Hint 1
Let the given facts prune the tree: Use what you already know (a large Tier-1 city, every order type affected, a start last week) to set aside whole groups of causes, then split each delivery into its stages to see which one grew.
Follow-up questions
- The data shows that the time from packing to rider pickup grew while travel time stayed flat. What do you investigate next?
- If the cause is heavy rain and waterlogging in the city, what would you change, and would you adjust the delivery time shown to customers?
- How would you tell whether the slower deliveries are already hurting repeat orders, and which metric would you watch?
Solution walkthrough is for members
Get the full walkthrough with reference code and complexity analysis, plus Kitty Coach at 300 messages a day. Hints, the code runner and community answers stay free.