Dataford
Popular roles
Software EngineerData AnalystData ScientistData EngineerBusiness AnalystAI EngineerMachine Learning EngineerProduct Manager
Browse
Browse All RolesEvery role hub, from analyst to MLBrowse All CompaniesCompany-specific interview loopsAll Interview GuidesThe full guide library
Top questions by role
Software EngineerData AnalystData ScientistData EngineerBusiness AnalystAI EngineerMachine Learning EngineerProduct Manager
Top questions by skill
SQLPythonStatisticsMachine LearningA/B TestingSystem DesignGenerative AIProduct SenseMetricsBehavioral
Browse all questions →Try a mock interview
ExperiencesCoaching
New
Practice
Mock InterviewsTimed interview simulations with feedbackSuccess PathYour 6-week structured planModulesCurated lessons by topicWebinarsTalks from ex-Big Tech data leadsPlaygroundA free-form scratch editor
Learn
BlogInterview strategy and career adviceTech Job Market ReportHiring trends across data and AI rolesFor UniversitiesDataford for career centersAbout DatafordWho we are and how we build
Pricing
Get started

Minimum Sample Size for AI Feature

MediumStatistics & Probability00:00
Practice interviewer
In session
5 left
00:00

Whenever you're ready, walk me through how you'd approach this.

You need to log in / sign up to chat or submit.

Problem

Business Context

NovaAssist is testing a new AI reply-suggestion feature in its customer support product. Traffic is limited because only a small share of agents currently use the beta workflow, so the team needs to know whether a standard A/B test is feasible before launch.

Problem Statement

Estimate the minimum sample size needed to detect a meaningful improvement in adoption rate, then assess whether the available traffic is enough to run the experiment in a reasonable time.

Given Data

The primary metric is agent adoption rate: the share of eligible sessions where the agent accepts at least one AI suggestion.

MetricValue
Baseline adoption rate18.0%
Minimum detectable effect (absolute lift)3.0 percentage points
Target treatment adoption rate21.0%
Significance level0.05
Desired power80%
Eligible sessions per week8,400
Planned traffic split50% control / 50% treatment

Requirements

  1. State the null and alternative hypotheses for the experiment.
  2. Compute the minimum required sample size per group for a two-sided test of two proportions.
  3. Compute the total required sample size.
  4. Estimate how many weeks NovaAssist would need to run the test with current traffic.
  5. Conclude whether the test is practical and name at least two alternatives if traffic is too limited.

Assumptions

  • Sessions are independently randomized at the eligible-session level.
  • The baseline rate of 18.0% is stable during the test window.
  • Use a normal approximation for sample size planning.
  • Ignore attrition and sample-ratio mismatch for the initial calculation.