PracHub
QuestionsPremiumCoachesLearningGuidesInterview Prep
|Home/Analytics & Experimentation/TikTok

Determine Metrics for Evaluating Homepage Recommendation Carousel

Last updated: Mar 29, 2026

Quick Overview

This question evaluates a data scientist's competency in designing and analyzing A/B experiments for a homepage recommendation carousel, including selection of primary and secondary metrics, assessment of statistical power and practical significance, and the setup of guardrails.

  • medium
  • TikTok
  • Analytics & Experimentation
  • Data Scientist

Determine Metrics for Evaluating Homepage Recommendation Carousel

Company: TikTok

Role: Data Scientist

Category: Analytics & Experimentation

Difficulty: medium

Interview Round: Onsite

##### Scenario Product team launches an A/B test introducing a new recommendation carousel on the homepage. ##### Question What primary and secondary metrics would you track to evaluate the experiment? Describe the decision framework you would use to determine whether to roll the feature out to all users. ##### Hints Discuss click-through rate, conversion, guardrails, power, practical significance.

Quick Answer: This question evaluates a data scientist's competency in designing and analyzing A/B experiments for a homepage recommendation carousel, including selection of primary and secondary metrics, assessment of statistical power and practical significance, and the setup of guardrails.

Related Interview Questions

  • Define Ultra success metrics and detect suspicious transactions - TikTok (easy)
  • Plan DS approach for biker delivery project - TikTok (easy)
  • Define and critique a user activity metric - TikTok (easy)
  • Design and decompose Trust & Safety risk metrics - TikTok (easy)
  • Analyze promo anomaly and design risk guardrails - TikTok (Medium)
TikTok logo
TikTok
Aug 4, 2025, 10:55 AM
Data Scientist
Onsite
Analytics & Experimentation
0
0

Experiment Evaluation: Homepage Recommendation Carousel

Scenario

A product team has shipped an A/B test for a new recommendation carousel placed on the app's homepage. Users are randomly assigned at the user level to Control (no carousel) or Treatment (homepage shows the carousel). The goal is to improve user engagement and downstream conversions without harming overall experience or performance.

Task

  1. Propose the primary metric(s) and secondary metric(s) you would track to evaluate the experiment.
  2. Describe a clear decision framework to determine whether to roll the feature out to all users, including how you will handle statistical power, practical significance, and guardrails.

Assume a standard 50/50 split, sticky assignment, and at least one weekly cycle of traffic.

Solution

Show

Submit Your Answer to Earn 20XP

Sign in to leave a comment

Loading comments...

Browse More Questions

More Analytics & Experimentation•More TikTok•More Data Scientist•TikTok Data Scientist•TikTok Analytics & Experimentation•Data Scientist Analytics & Experimentation
PracHub

Master your tech interviews with 8,000+ real questions from top companies.

Product

  • Questions
  • Learning Tracks
  • Interview Guides
  • Resources
  • Premium
  • For Universities
  • Student Access

Browse

  • By Company
  • By Role
  • By Category
  • Topic Hubs
  • SQL Questions
  • Compare Platforms
  • Discord Community

Support

  • support@prachub.com
  • (916) 541-4762

Legal

  • Privacy Policy
  • Terms of Service
  • About Us

© 2026 PracHub. All rights reserved.