##### Scenario You are tasked with building a model that predicts user churn for a subscription app. ##### Question Which modeling techniques would you consider and why? How would you address class imbalance and choose appropriate evaluation metrics? ##### Hints Talk about logistic regression, tree models, resampling, ROC-AUC, precision-recall.

This question evaluates skills in predictive modeling for churn, temporal feature engineering and data-leakage awareness, handling class imbalance, and selecting evaluation metrics for subscription retention scenarios in the Machine Learning domain.

How do I approach Machine Learning interview questions?

Machine Learning questions require understanding of core concepts and practice. PracHub provides solutions with explanations to help you master machine learning interviews.

What difficulty level is this interview question?

This is a medium difficulty Machine Learning question, commonly asked during Onsite rounds at TikTok.

What role is this question designed for?

This question is commonly asked for Data Scientist candidates at TikTok during technical interviews.

Predict User Churn with Effective Modeling Techniques

Predicting User Churn for a Subscription App

Context

You are building a model to predict which active subscribers are likely to churn soon so the team can target retention offers. Assume:

Label: churn in the next 30 days (cancel subscription or no activity for 30 days).
Features: recent engagement (recency, frequency, session duration), tenure, plan type, payment history, support interactions, device/geo, marketing touches.
Data is time-ordered; avoid leakage by using only information available before the prediction date.

Tasks

(a) Which modeling techniques would you consider and why?

(b) How would you address class imbalance?

Predicting User Churn for a Subscription App

Context

You are building a model to predict which active subscribers are likely to churn soon so the team can target retention offers. Assume:

Label: churn in the next 30 days (cancel subscription or no activity for 30 days).
Features: recent engagement (recency, frequency, session duration), tenure, plan type, payment history, support interactions, device/geo, marketing touches.
Data is time-ordered; avoid leakage by using only information available before the prediction date.

Tasks

(a) Which modeling techniques would you consider and why?

(b) How would you address class imbalance?

Predict User Churn with Effective Modeling Techniques

Quick Overview

Predicting User Churn for a Subscription App

Context

Tasks

Solution

Submit Your Answer to Earn 20XP

Predict User Churn with Effective Modeling Techniques

Quick Overview

Predicting User Churn for a Subscription App

Context

Tasks

Solution

Submit Your Answer to Earn 20XP