Chime: Design and Analyze A/B Test for Recommendation Widget

Quick Overview

This interview question evaluates metric design, causal reasoning, experiment setup, diagnostics, SQL/statistical checks, and recommendations in a realistic interview setting. A strong answer for Design and Analyze A/B Test for Recommendation Widget states assumptions, handles edge cases, explains trade-offs, and shows how to validate the result clearly.

Scenario

You are designing and analyzing an online A/B test for launching a new recommendation widget in a consumer-facing product (e.g., mobile and web app). The widget recommends relevant actions or products on a home/feed surface.

Task

Explain, end-to-end, how you would set up, run, and analyze an A/B test for this recommendation widget.

Requirements

Experiment design and randomization
- Define hypothesis, unit of assignment, eligibility/exposure, and rollout plan.
Sample size and power
- Determine minimum detectable effect (MDE), sample size, duration, and traffic ramp.
Metrics
- Choose a single primary metric, key secondary metrics, and guardrail metrics; define how each is computed.
Execution and debugging
- Instrumentation, logging, pre-checks (e.g., SRM), and live monitoring.
Analysis
- Statistical tests, variance reduction, handling triggered exposure vs ITT, and multiple comparisons.
Pitfalls
- List issues that could invalidate the experiment and how you would detect/mitigate them.
Communication
- How you would communicate results and a go/no-go recommendation to stakeholders.

Assume users can be exposed multiple times across sessions and platforms, and there is no cross-user network effect.

Clarifying Questions to Ask Guidance

Clarify the business objective, unit of analysis, time window, exposure definition, and primary metric.
State assumptions about instrumentation, randomization, sample size, and data quality.
Separate descriptive analysis from causal claims.

What a Strong Answer Covers Guidance

A metric framework with primary, guardrail, and diagnostic metrics.
A credible analysis or experiment design with clear assumptions and bias checks.
SQL/statistical logic for segmentation, variance, confidence, and data validation where relevant.
An actionable recommendation that explains trade-offs and next steps.

Follow-up Questions Guidance

What sanity checks would you run before trusting the result?
How would you handle novelty effects, seasonality, or selection bias?
What decision would you make if metrics disagree?

Design and Analyze A/B Test for Recommendation Widget

Quick Overview

Design and Analyze A/B Test for Recommendation Widget

Design and Analyze A/B Test for Recommendation Widget

Scenario

Task

Requirements

Clarifying Questions to Ask Guidance

What a Strong Answer Covers Guidance

Follow-up Questions Guidance

Write your answer