← Amazon Interview Insights

Amazon·Data Scientist·Technical Phone Screen·Senior

Senior
May 2026

Summary

Amazon DS interview, product/experimentation heavy. One big open-ended case about monetization for a mobile game that basically ate the whole session. Felt like a strategy + stats hybrid and I wasn't fully prepared for how deep they wanted to go on the experiment design side.

Questions Asked (1)

Q1

You're a data scientist at a free-to-play mobile game company. The product team wants to add video ads (rewarded and/or interstitial). Design a full evaluation plan covering success metrics, experiment design, incremental impact estimation, threats to validity, and a ship/don't ship framework.

A/B Testing & ExperimentationPricing & MonetizationProduct Analytics & Metrics
Author's notes

This was the whole interview basically.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by clarifying the business goal (e.g., maximize long-term player LTV) and the ad types (rewarded vs. interstitial). Then outline a structured plan covering metrics, experiment design, incremental impact estimation, validity threats, and a ship/don't ship decision framework. Emphasize the importance of measuring incremental impact and considering long-term effects.

Pro tip: Propose a long-term holdout group to measure the persistent effects of ads on retention and monetization, and use a difference-in-differences approach to isolate the incremental impact from other factors.

1. Define Success Metrics

Identify primary metrics (e.g., incremental revenue per user, retention) and guardrail metrics (e.g., player satisfaction, churn). Consider both short-term and long-term effects.

2. Design the Experiment

Choose a randomized controlled trial with a control group (no ads) and treatment groups (rewarded, interstitial, or both). Determine sample size, duration, and randomization unit (e.g., user-level).

3. Estimate Incremental Impact

Use methods like difference-in-differences or causal inference to measure the incremental effect of ads on key metrics, accounting for seasonality and external factors.

4. Identify Threats to Validity

Consider novelty effects, cannibalization of in-app purchases, selection bias, and network effects. Plan sensitivity analyses and holdout groups to mitigate.

5. Ship/Don't Ship Framework

Define decision criteria based on statistical significance, practical significance (effect size), and alignment with long-term strategy. Include a cost-benefit analysis and stakeholder alignment.

Key Points to Mention

  • Incremental impact measurement using control groups and difference-in-differences
  • Long-term holdout to detect persistent effects on retention and monetization
  • Guardrail metrics to ensure player experience isn't harmed
  • Cannibalization of in-app purchases and potential revenue shifts
  • Novelty effect and how to account for it in experiment duration
  • Statistical power and sample size calculation for reliable results

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