← DoorDash Interview Insights

DoorDash·Data Scientist·Technical Phone Screen·Senior

Senior
Jun 2026

Summary

DoorDash data science interview focused almost entirely on a deep experiment design case for the notifications team. It was one of the more technically demanding sessions I've had, covering everything from randomization strategy to long-term holdout design in a single extended problem.

Questions Asked (5)

Q1

How would you design an experiment to measure the incremental impact of increasing push notification frequency and/or using smarter send timing for order-related notifications? Walk through your randomization unit, exposure caps, time-of-day stratification, and how you'd control for interference from other channels like email and SMS.

A/B Testing & ExperimentationProduct Analytics & Metrics
Author's notes

This is where I spent most of my time.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by framing the experiment as a multi-factor test to isolate the incremental impact of frequency and timing, then detail the randomization unit (e.g., user-level) and exposure caps to avoid user fatigue. Walk through stratification by time zones and control for channel interference by measuring cross-channel effects and using holdout groups.

Pro tip: Emphasize the importance of pre-registering the analysis plan and using intent-to-treat (ITT) analysis to avoid bias from non-compliance, especially when users can opt out of notifications.

1. Define hypotheses and metrics

Clearly state the null and alternative hypotheses for frequency and timing, and select primary (e.g., incremental orders) and guardrail metrics (e.g., unsubscribes, app opens).

2. Choose randomization unit and design

Randomize at the user level to avoid contamination, and consider a factorial design (2x2) to test frequency and timing independently and their interaction.

3. Implement exposure caps and stratification

Set daily/weekly caps on notifications per user to prevent fatigue, and stratify randomization by time zone and user activity level to balance diurnal patterns.

4. Control for cross-channel interference

Include email and SMS in the analysis by measuring their send volumes and using a holdout group that receives no order-related notifications to estimate the incremental effect.

5. Analyze and validate

Use ITT analysis with regression adjustment for covariates, check for novelty effects, and run sensitivity analyses to ensure robustness.

Key Points to Mention

  • User-level randomization to prevent spillover effects
  • Factorial design to test frequency and timing simultaneously
  • Exposure caps to mitigate user fatigue and notification opt-outs
  • Time-of-day stratification to account for diurnal patterns
  • Cross-channel interference: measure email/SMS sends and use holdout groups
  • Intent-to-treat analysis and pre-registration to avoid p-hacking

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

Q2

What are your primary success metrics and guardrails for this notification experiment, and how do you aggregate them across both per-user and per-notification dimensions?

Product Analytics & MetricsA/B Testing & Experimentation
Author's notes

Pretty natural territory for me.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by clarifying the experiment's goal (e.g., increasing order frequency or engagement) and then define primary success metrics (e.g., orders per user, click-through rate) and guardrail metrics (e.g., unsubscribes, complaint rate). Explain how you aggregate metrics at both per-user and per-notification levels, ensuring you account for multiple notifications per user and avoid metric dilution.

Pro tip: Emphasize that guardrails should be monitored at the per-user level to detect user-level harm, while success metrics may be aggregated per-notification for granular insights; always check for novelty effects and use holdout groups for long-term validation.

1. Clarify Experiment Objective

Restate the goal of the notification experiment (e.g., drive incremental orders, improve retention) to align metrics with business outcomes.

2. Define Success Metrics

Select primary success metrics (e.g., orders per user, notification click-through rate) and secondary metrics that capture the desired behavior change.

3. Identify Guardrail Metrics

Choose guardrails that protect user experience and platform health (e.g., unsubscribe rate, notification disablement, complaint rate, app uninstalls).

4. Aggregate Across Dimensions

Explain how to compute metrics at per-user level (e.g., average orders per user) and per-notification level (e.g., CTR per notification), and how to handle users receiving multiple notifications (e.g., weighting, clustering).

5. Monitor and Iterate

Describe how you would monitor metrics over time, check for statistical significance, and decide whether to roll out, iterate, or stop the experiment.

Key Points to Mention

  • Primary success metrics: orders per user, notification click-through rate, conversion rate
  • Guardrail metrics: unsubscribe rate, notification disablement, complaint rate, app uninstalls
  • Per-user aggregation: average per user, using user-level randomization to avoid bias
  • Per-notification aggregation: CTR, conversion rate per notification, accounting for multiple notifications per user
  • Statistical methods: clustered standard errors, mixed-effects models, or bootstrap to handle multiple notifications per user
  • Novelty effect and long-term holdout to validate sustained impact

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

Q3

How would you measure the long-term effects of increased notification frequency, including novelty effects, user fatigue, and whether any lift you see in the first couple weeks actually persists?

A/B Testing & ExperimentationProduct Strategy
Author's notes

Honestly the part I was least prepared for.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by acknowledging that short-term A/B tests can be misleading due to novelty effects and user fatigue, so you need a long-term holdout group and extended measurement. Then outline a plan to track both behavioral metrics (e.g., open rates, opt-outs) and business metrics (e.g., retention, orders) over several weeks, using techniques like cohort analysis and time-series decomposition to separate novelty from sustained impact.

Pro tip: Propose a 'holdout' group that never receives the increased frequency, and measure the difference-in-differences over time to isolate the true long-term effect from novelty. Also, consider running a switchback or staggered rollout to account for external factors.

1. Define long-term success metrics

Identify metrics that capture both user engagement (e.g., notification open rate, click-through rate) and user well-being (e.g., opt-out rate, app uninstalls, notification disablement). Also include business outcomes like retention, order frequency, and customer lifetime value.

2. Design experiment with long-term holdout

Set up an A/B test with a control group that receives the current frequency and a treatment group with increased frequency. Include a long-term holdout group that never receives the increased frequency to measure the cumulative effect over months.

3. Measure novelty and fatigue effects

Track metrics over time (e.g., weekly) to observe initial lift (novelty) and subsequent decline (fatigue). Use time-series analysis or cohort analysis to separate short-term spikes from sustained changes.

4. Analyze persistence of effects

Compare treatment vs. control after the novelty period (e.g., after 4-6 weeks) to see if any lift persists. Use difference-in-differences or regression discontinuity to control for confounders.

5. Consider external validity and segment analysis

Check if effects vary by user segment (e.g., new vs. existing users, high vs. low engagement). Also, ensure the experiment duration covers enough time to capture long-term behavior changes.

Key Points to Mention

  • Novelty effect: initial spike in engagement due to newness, which may fade.
  • User fatigue: increased frequency can lead to annoyance, opt-outs, or disengagement.
  • Long-term holdout group to measure cumulative impact.
  • Difference-in-differences or time-series analysis to isolate treatment effect.
  • Segment analysis to identify heterogeneous treatment effects.
  • Business metrics like retention and LTV, not just engagement metrics.

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

Q4

Users in this experiment receive multiple notifications over the study period. How do you handle the repeated exposure problem and model diminishing returns or saturation effects in your analysis?

A/B Testing & ExperimentationData Modeling
Author's notes

I talked through dose-response curves, basically binning users by number of notifications received and plotting the outcome metric across bins.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Acknowledge that repeated notifications create non-independent treatment exposures and potential saturation, then propose modeling the dose-response relationship using per-user notification counts and time-varying covariates. Suggest methods like survival analysis, marginal structural models, or generalized additive models to capture diminishing returns, and validate with sensitivity analyses.

Pro tip: Frame the problem as a dose-response curve rather than a simple A/B test, and emphasize that the goal is to find the optimal notification frequency that maximizes long-term user engagement without causing fatigue.

1. Define the exposure metric

Quantify each user's treatment exposure as the cumulative number of notifications received, and consider time since first notification to capture dynamic effects.

2. Model the dose-response relationship

Use regression splines, polynomial terms, or non-linear models to estimate how the outcome changes with increasing notification count, allowing for saturation.

3. Account for time-varying confounding

Apply marginal structural models with inverse probability weighting to adjust for time-varying factors that affect both notification delivery and user behavior.

4. Incorporate user heterogeneity

Use mixed-effects models or latent class analysis to capture individual differences in responsiveness and saturation thresholds.

5. Validate and interpret

Perform sensitivity analyses (e.g., different functional forms, negative controls) and translate the modeled curve into actionable insights like optimal send frequency.

Key Points to Mention

  • Repeated exposure violates the stable unit treatment value assumption (SUTVA) and can bias naive A/B test estimates.
  • Diminishing returns can be modeled with concave functions (e.g., logarithmic, square root) or splines.
  • Saturation effects may lead to negative marginal utility; consider piecewise models or change-point detection.
  • Time-varying confounding requires methods like marginal structural models or g-computation.
  • User-level random effects or clustering can account for correlation among repeated observations.
  • Practical implications: determine optimal notification cadence and identify user segments prone to fatigue.

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

Q5

Walk through your power and MDE assumptions for this experiment, your stopping rules, and how you'd build a decision framework that weighs short-term conversion lift against potential long-term retention damage. What's your concrete rollback criterion?

A/B Testing & ExperimentationProduct StrategyTechnical Trade-offs
Author's notes

I anchored MDE around something like a 1 to 2 percent relative lift on incremental orders, which felt realistic for a notification change.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Structure your answer by first stating the business context and the key metric (e.g., conversion) with its baseline and MDE, then explain how you set power and stopping rules, and finally present a decision framework that incorporates leading indicators of long-term retention. Emphasize the trade-off and provide a concrete rollback criterion based on guardrail metrics.

Pro tip: Show that you think beyond statistical significance by including practical significance and business impact; mention that you'd pre-register the decision framework and rollback criteria to avoid post-hoc rationalization.

1. Define metrics and assumptions

State the primary metric (e.g., conversion rate) and its baseline, the minimum detectable effect (MDE) you care about, and the guardrail metrics for retention (e.g., 7-day retention). Specify power (80%) and significance level (5%).

2. Set stopping rules

Explain that you'll use a fixed-horizon test or sequential testing with alpha-spending to control false positives. Mention that you'll monitor guardrails continuously and stop early only if guardrails are breached.

3. Build decision framework

Create a 2x2 matrix: primary metric lift vs. guardrail impact. Define thresholds for shipping, iterating, or rolling back. Incorporate leading indicators of long-term retention (e.g., repeat order rate) if available.

4. Define rollback criterion

Specify a concrete threshold: e.g., if 7-day retention drops by more than 1% relative and is statistically significant, roll back immediately. Also consider practical significance and business impact.

5. Communicate and iterate

Explain how you'd communicate the trade-off to stakeholders and propose next steps, such as a follow-up experiment with a modified treatment to mitigate retention damage.

Key Points to Mention

  • Baseline conversion rate and MDE calculation (e.g., using power analysis)
  • Statistical power (80%) and significance level (5%)
  • Stopping rules: fixed horizon vs. sequential testing, alpha spending
  • Guardrail metrics: retention, churn, customer lifetime value
  • Decision framework: ship, iterate, or rollback based on primary and guardrail metrics
  • Concrete rollback criterion: e.g., retention drop >1% relative and statistically significant

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