Build a metric tree for Dasher App after a major redesign at global scale

Problem Statement Description

Product context: DoorDash is a local commerce and delivery platform; its products include restaurant delivery, DashPass, grocery and retail delivery, merchant tools, and dasher tools.

DoorDash has rolled out a major redesign of the Dasher App across multiple markets. The app is central to how Dashers receive offers, decide whether to accept them, navigate to restaurants, manage pickup issues, complete deliveries, track earnings, and interact with support. Because the redesign touches core workflow surfaces, leadership needs a metric tree that can evaluate whether the new experience improves Dasher productivity and marketplace reliability without harming restaurants, consumers, or DoorDash unit economics.

Your task is to define how DoorDash should measure the impact of this redesign at global scale. The metric tree should connect high-level business and marketplace outcomes to the specific Dasher behaviors and product interactions that the redesigned app is expected to influence. It should also account for differences across geographies, delivery modes, restaurant operations, order complexity, and Dasher tenure.

This is a metrics interview question, so focus on what should be measured, how metrics should be structured, and how the metric tree would help teams diagnose whether the redesign is working. Avoid jumping directly to product fixes; instead, make the measurement system clear enough that product, operations, engineering, and marketplace teams could use it to make rollout and iteration decisions.

The experience should consider:

- The primary success metric and the denominator that makes it meaningful, such as per order, per active Dasher hour, per delivery offer, or per completed delivery.

- Key funnel stages in the Dasher workflow, including offer receipt, acceptance, travel to restaurant, pickup, delivery completion, and post-delivery actions.

- Instrumentation requirements for redesigned surfaces, including exposure logging, event quality, latency, app version, market, device, and session context.

- Cohorts and cuts that matter, such as new vs. experienced Dashers, market maturity, restaurant wait-time patterns, batched orders, peak vs. off-peak periods, and delivery verticals.

- Guardrail metrics for consumer experience, restaurant impact, Dasher earnings, safety, cancellations, support contacts, fraud, and marketplace balance.

- Diagnostic metrics that distinguish app usability issues from external factors like restaurant preparation delays, weather, demand spikes, or supply shortages.

- Decision usefulness for staged rollout, experiment readouts, rollback triggers, and ongoing monitoring after global launch.

The goal is to produce a metric tree that gives DoorDash a clear, actionable view of whether the Dasher App redesign improves the delivery marketplace while protecting reliability, Dasher trust, restaurant operations, and long-term business health.