Root cause a sudden decline in retention among restaurants using Dasher App 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 observed a sudden decline in retention among restaurant partners whose order fulfillment depends on Dasher App workflows. The drop appears at global scale and may affect restaurants’ willingness to continue relying on DoorDash for delivery, pickup coordination, and marketplace demand. Your task is to investigate the issue as a product leader and structure a root-cause analysis.

Assume restaurant retention reflects whether merchants continue actively fulfilling orders through DoorDash over a defined period. The decline could be related to restaurant operations, Dasher pickup behavior, app changes, marketplace conditions, quality issues, incentives, competition, seasonality, or measurement problems. You should avoid jumping to a single cause and instead define how you would isolate where, when, and for whom the retention decline is happening.

This is a hard RCA question because it spans a three-sided marketplace: restaurants, Dashers, and consumers. Changes in the Dasher App can indirectly affect restaurants through pickup delays, cancellations, missed handoffs, order accuracy issues, merchant wait times, support burden, or degraded merchant ROI. Your investigation should account for global market differences, operational dependencies, and the possibility that the observed retention decline is caused by instrumentation or definition changes rather than true merchant churn.

The experience should consider:

- How restaurant retention is defined, including denominator, activity threshold, lookback window, and whether the drop is in logo retention, order volume retention, or fulfillment retention.

- Segmentation by geography, cuisine type, restaurant size, marketplace maturity, delivery mode, order volume tier, integration type, and new versus tenured merchants.

- Timeline analysis around product releases, Dasher App changes, dispatch logic updates, pricing/incentive changes, support policy changes, outages, or external events.

- Instrumentation checks to confirm whether tracking, event schemas, merchant identifiers, or retention calculations changed.

- Marketplace health indicators such as Dasher supply, pickup wait time, cancellations, lateness, acceptance/completion rates, restaurant complaints, refunds, and consumer demand.

- Evidence needed to validate or reject hypotheses, including cohort comparisons, funnel analysis, experiment readouts, merchant feedback, support tickets, and operational dashboards.

- Immediate mitigation options if restaurants are at risk, balanced against Dasher earnings, consumer experience, delivery reliability, and unit economics.

- Longer-term prevention mechanisms such as monitoring, alerting, release gates, and cross-functional ownership across product, operations, analytics, merchant success, and engineering.

Your goal is to present a structured RCA approach that narrows the anomaly, separates true behavioral change from measurement error, identifies the most likely drivers, and outlines how DoorDash should respond without compromising marketplace balance.