DoorDash·Data Scientist·Technical Phone Screen·Senior

SeniorPrefer not to say

Jun 2026

Summary

A data science case interview at DoorDash focused entirely on one deep, open-ended question about diagnosing a drop in completed delivery orders in LA. It was the kind of question that sounds manageable until you're twenty minutes in and realize you've only covered half of what they wanted.

Questions Asked (1)

Completed delivery orders in Los Angeles have dropped noticeably over the past two weeks compared to historical trends and similar cities. Walk me through how you'd investigate this end-to-end, from confirming the drop is real all the way to proposing an experiment to test a fix.

Root Cause AnalysisProduct Analytics & MetricsA/B Testing & Experimentation

Author's notes

This one ate up the whole session.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by validating the data and ruling out instrumentation or pipeline issues, then segment the drop by geography, time, user, and order characteristics to isolate the cause. Form hypotheses about potential drivers (supply, demand, competition, product changes), test them with data, and finally design an experiment to validate the most promising fix.

Pro tip: Always consider both supply-side (Dasher availability) and demand-side (consumer behavior) factors, and check for external events like weather, holidays, or competitor promotions that could explain the drop.

1. Validate the Drop

Confirm the drop is real by checking data quality, pipeline integrity, and comparing with other metrics like app traffic or order attempts. Ensure it's not a reporting artifact.

2. Segment and Localize

Break down the drop by dimensions such as time (daily/hourly), geography (neighborhoods), user segments (new vs. existing), and order characteristics (restaurant type, delivery time) to identify where the drop is concentrated.

3. Generate Hypotheses

Brainstorm potential causes: supply issues (Dasher shortage), demand shifts (consumer behavior), competitive actions, product changes, or external events. Prioritize based on data signals.

4. Test Hypotheses with Data

Use statistical analysis and causal inference methods (e.g., difference-in-differences, regression) to test each hypothesis. Look for correlations and potential causal links.

5. Design an Experiment

Propose an A/B test to validate the fix for the most likely cause. Define success metrics, randomization unit, sample size, and duration. Consider potential confounders.

Key Points to Mention

Data validation: check for logging errors, pipeline delays, or metric definition changes.
Segmentation: analyze by geography, time, user cohorts, and order types to localize the issue.
Supply vs. demand: investigate Dasher supply (e.g., active Dashers, acceptance rates) and consumer demand (e.g., app opens, search queries).
External factors: consider weather, holidays, local events, or competitor promotions.
Causal inference: use methods like difference-in-differences or regression discontinuity to establish causality.
Experiment design: outline A/B test with clear hypothesis, metrics, randomization, and power analysis.

AI-generated suggestions, not part of the candidate's original notes. May be inaccurate — verify before relying on them.