Modify Instagram Feature: Track User Engagement Metric

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 Modify Instagram Feature: Track User Engagement Metric states assumptions, handles edge cases, explains trade-offs, and shows how to validate the result clearly.

Modify Instagram Feature: Track User Engagement Metric

Company: Yelp

Role: Data Scientist

Category: Analytics & Experimentation

Difficulty: medium

Interview Round: Technical Screen

##### Scenario A social-media company is considering modifying a disliked Instagram-like feature and must run an A/B test before shipping the change. ##### Question Pick a popular consumer app (e.g., Instagram) and name a feature you do not like. What single or composite metric would you track to measure user response? How would you estimate the required sample size and decide how long the test should run? State your assumptions and calculations. Which user segment or geography would you target first and why? What statistical test would you apply to compare control and treatment groups and why is it appropriate? Suppose the overall test is not statistically significant but the treatment group shows higher engagement. How would you interpret this outcome? Given the above, would you (a) fully launch the change, (b) rerun or extend the experiment, or (c) dig into a specific subgroup such as younger users? Justify your choice. ##### Hints Discuss metric choice, power analysis, duration, segmentation, test selection, and post-test decisions.

Quick Answer: 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 Modify Instagram Feature: Track User Engagement Metric states assumptions, handles edge cases, explains trade-offs, and shows how to validate the result clearly.

|Home/Analytics & Experimentation/Yelp
Yelp logo
Yelp
Aug 4, 2025, 10:55 AM
mediumData ScientistTechnical ScreenAnalytics & Experimentation
91
0

Modify Instagram Feature: Track User Engagement Metric

A/B Test Design for Modifying a Disliked Social Feature

Scenario

A social-media company plans to modify a disliked Instagram-like feature and must run an A/B test before shipping the change.

Task

Pick a popular consumer app and name a feature you do not like. Then answer the following:

  1. What single or composite metric would you track to measure user response? State why it aligns with the product goal.
  2. How would you estimate the required sample size and decide how long the test should run? Show assumptions and calculations.
  3. Which user segment or geography would you target first and why?
  4. What statistical test would you apply to compare control and treatment groups and why is it appropriate?
  5. Suppose the overall test is not statistically significant but the treatment shows higher engagement. How would you interpret this outcome?
  6. Given the above, would you (a) fully launch the change, (b) rerun or extend the experiment, or (c) dig into a specific subgroup (e.g., younger users)? Justify your choice.

Hints: Discuss metric choice, power analysis, duration, segmentation, test selection, and post-test decisions.

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?
Loading comments...