← Uber Interview Insights

Uber·Product Manager·Onsite - Product Sense / Strategy·Senior

Senior
Jun 2026

Summary

PM interview at Uber, one question about experimentation design. Pretty standard but the details matter more than you'd think.

Questions Asked (1)

Q1

How would you design an A/B test for a new campaign?

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

I jumped straight to metric selection and kind of glossed over the setup stuff like randomization unit and sample size.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by clarifying the campaign's goal and the key metric you want to move, then outline a structured A/B test plan covering hypothesis, randomization, sample size, and success criteria. Emphasize how you would ensure validity and interpret results to make a data-driven decision.

Pro tip: Highlight the importance of defining a clear primary metric and guardrail metrics upfront, and mention how you would handle common pitfalls like novelty effects or network effects, especially in a marketplace like Uber.

1. Define Objective and Hypothesis

Clarify the campaign's goal (e.g., increase conversions, engagement) and state a testable hypothesis about how the change will impact a specific metric.

2. Design the Experiment

Determine the control and treatment groups, randomization unit (e.g., user, session), and ensure proper sample size and power calculation to detect a meaningful effect.

3. Select Metrics and Guardrails

Choose a primary success metric, secondary metrics, and guardrail metrics to monitor for unintended negative consequences.

4. Run the Test and Monitor

Launch the test, monitor for data quality, and ensure no external factors bias the results; avoid peeking at results prematurely.

5. Analyze and Decide

After the test duration, analyze results for statistical significance, practical significance, and segment-level insights; decide whether to roll out, iterate, or abandon.

Key Points to Mention

  • Randomization and avoiding selection bias
  • Sample size and statistical power
  • Primary, secondary, and guardrail metrics
  • Statistical significance vs. practical significance
  • Potential pitfalls: novelty effect, network effects, seasonality
  • Iterative testing and learning

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