Create a metric tree for repeat group orders in food-delivery

Problem Statement Description

You are a PM working on Swiggy’s group ordering experience for office lunches in urban India. The target user is typically an office lunch coordinator who starts or manages a shared order for a team, gets colleagues to participate, handles timing and preferences, and expects the order to arrive reliably during a limited lunch window.

The interview task is to create a metric tree for repeat group orders. The focus is not just on whether a group order happened once, but whether the same coordinator, team, workplace, or participant group comes back to use the experience again. Your metric structure should help Swiggy understand what drives repeat behavior and where the funnel or experience may be breaking down.

The metric tree should account for the realities of food delivery: restaurant availability, prep-time variability, delivery batching, payment coordination, cancellations, substitutions, late participants, and trust in delivery accuracy. It should also be useful for product decisions, experimentation, and operational diagnosis rather than being a list of disconnected numbers.

The experience should consider:

- How to define a “repeat group order” clearly, including the actor, time window, and denominator.

- Whether repeat behavior should be measured at coordinator, office, group, participant, or restaurant level.

- The full group-order workflow: creation, invite sharing, participant joining, cart building, payment, checkout, delivery, and post-order satisfaction.

- Instrumentation needed to distinguish individual orders from true group orders and to capture drop-offs accurately.

- Cohorts such as new vs returning coordinators, office size, weekday lunch patterns, city, restaurant type, and delivery distance.

- Guardrail metrics around delivery reliability, cancellation, refunds, customer support contacts, restaurant load, and delivery partner efficiency.

- Data quality, privacy, and consent considerations when associating users with workplaces or recurring groups.

- How the metric tree would help diagnose whether repeat orders are limited by demand, product friction, supply availability, or operational reliability.

The goal is to define a clear, decision-useful metric hierarchy that connects the business outcome of repeat group orders to measurable user behaviors, funnel steps, quality signals, and operational constraints in Swiggy’s food-delivery marketplace.