← Amazon Interview Insights

Amazon·Data Scientist·Technical Phone Screen·Senior

Senior
May 2026

Summary

Amazon data scientist case study focused entirely on pricing strategy for a subscription AI product. One big open-ended prompt covering analysis, experiment design, and segmentation. Pretty demanding for a single question.

Questions Asked (1)

Q1

You work on a subscription-based AI video editing product and leadership wants to raise prices. How would you analyze whether the increase makes sense, decide what change to ship and for whom, and design an experiment to measure impact and inform a launch decision?

Pricing & MonetizationA/B Testing & ExperimentationProduct Analytics & Metrics
Author's notes

This is a lot to hold in your head at once.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by clarifying the business objective and constraints, then propose a data-driven framework that covers diagnosis, segmentation, experiment design, and decision criteria. Emphasize how you would balance short-term revenue gains with long-term customer retention and product value perception.

Pro tip: Anchor your analysis in customer willingness-to-pay and elasticity, and always include a holdout group to measure long-term effects—leadership will appreciate the rigor and foresight.

1. Diagnose current state and objectives

Understand why leadership wants to raise prices: revenue targets, cost pressures, or value enhancements. Analyze current pricing, customer segments, usage patterns, and competitive landscape to establish a baseline.

2. Estimate price elasticity and willingness to pay

Use historical data, surveys (e.g., Van Westendorp), and conjoint analysis to estimate how demand changes with price. Identify which segments are most and least price-sensitive.

3. Define the price change and target segments

Based on elasticity and business goals, decide on the magnitude and structure of the price increase (e.g., flat increase, tiered pricing) and which customer segments to apply it to (e.g., new vs. existing, heavy vs. light users).

4. Design and run a controlled experiment

Randomize eligible customers into treatment (new price) and control (old price) groups. Define primary metrics (e.g., conversion, retention, revenue per user) and guardrail metrics (e.g., churn, customer satisfaction). Ensure sufficient power and duration.

5. Analyze results and recommend launch decision

Compare treatment vs. control on key metrics, assess statistical significance, and evaluate long-term impact via holdout. Weigh revenue gains against churn risks and provide a clear go/no-go recommendation with rollout plan.

Key Points to Mention

  • Price elasticity of demand and methods to estimate it (e.g., regression, conjoint, Van Westendorp)
  • Customer segmentation and differential pricing strategies
  • A/B test design: randomization, sample size, power, duration, and guardrail metrics
  • Long-term impact measurement using holdout groups and cohort analysis
  • Competitive analysis and perceived value of AI features
  • Decision framework balancing revenue, retention, and customer lifetime value

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