← Google Interview Insights

Google·Data Scientist·Take-home Assignment·Senior

SeniorPrefer not to say
May 2026

Summary

Got hit with a massive take-home style case at Google for a DS role. Two giant sub-problems back to back: a full end-to-end pricing A/B test design for a B2B SaaS company, and a 2x2 factorial experiment for a signup CTA. The depth expected here was genuinely surprising, way beyond what I'd seen in typical DS interviews.

Questions Asked (6)

Q1

You join a B2B SaaS company with three pricing tiers. The PM wants to run a 2-week A/B test to raise list prices by 20% on the two lower tiers. Should you run a user-level A/B test or a quasi-experimental design? Lay out the pros and cons including contamination risk, fairness concerns, and operational overhead.

A/B Testing & ExperimentationPricing & MonetizationTechnical Trade-offs
Author's notes

This is where I spent most of my time and still felt like I left things on the table.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by clarifying the goal: to estimate the causal impact of a 20% price increase on conversion and revenue. Then compare user-level A/B testing and quasi-experimental designs across contamination risk, fairness, and operational overhead, ultimately recommending a hybrid or staged approach that balances rigor with practicality.

Pro tip: In B2B SaaS, pricing experiments often suffer from spillover effects because users within the same company can be in different tiers; consider randomizing at the account or firm level instead of the user level to avoid contamination and fairness issues.

1. Clarify the objective and constraints

Define the primary metric (e.g., conversion rate, revenue per user) and constraints such as the 2-week duration, sample size, and business risk tolerance. Understand that pricing changes can have long-term effects beyond the test window.

2. Evaluate user-level A/B testing

Assess pros: high internal validity, precise measurement of user-level behavior. Cons: contamination risk if users interact (e.g., sharing pricing info), fairness concerns if similar users see different prices, and operational overhead of implementing price variations.

3. Evaluate quasi-experimental designs

Consider designs like difference-in-differences, synthetic control, or interrupted time series. Pros: avoids contamination and fairness issues by applying price changes to entire markets or time periods. Cons: lower internal validity due to confounding factors, requires more data and careful modeling.

4. Compare on key dimensions

Systematically compare contamination risk, fairness, and operational overhead. User-level A/B tests have high contamination and fairness risks but lower operational overhead; quasi-experiments have lower contamination and fairness risks but higher operational overhead and analytical complexity.

5. Recommend a hybrid or staged approach

Propose a pragmatic solution: e.g., start with a quasi-experiment in a few markets to gauge impact, then if needed, run a cluster-randomized A/B test at the account level. Emphasize the need for stakeholder alignment and monitoring for unintended consequences.

Key Points to Mention

  • Contamination risk: users in the same company or social network may share pricing information, leading to spillover effects that bias results.
  • Fairness concerns: charging different prices to similar users can be perceived as unfair and damage trust, especially in B2B relationships.
  • Operational overhead: implementing user-level price variations requires engineering changes, customer support training, and legal review, which can be costly and slow.
  • Quasi-experimental designs: methods like difference-in-differences or synthetic control can provide causal estimates without randomization but require strong assumptions and more data.
  • Cluster randomization: randomizing at the account or firm level can mitigate contamination and fairness issues while retaining some experimental control.
  • Long-term effects: pricing changes may have lasting impacts on churn and brand perception, so consider running the test longer or using holdout groups.

AI-generated suggestions, not part of the candidate's original notes. May be inaccurate — verify before relying on them.

Q2

Define the full success metric hierarchy for this pricing experiment. What is your primary metric, what are the guardrails, and what exact decision threshold would you use to call the test a win?

A/B Testing & ExperimentationProduct Analytics & MetricsPricing & Monetization
Author's notes

I went with incremental LTV per acquired customer as the primary, which felt right.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by clarifying the experiment's goal and the business context, then define a metric hierarchy with a primary success metric, secondary metrics, and guardrail metrics. Specify a decision threshold that balances statistical significance, practical significance, and risk, and explain how you would handle multiple testing and novelty effects.

Pro tip: Emphasize that guardrails should be tied to long-term company health, not just short-term revenue, and propose a pre-registered analysis plan to avoid p-hacking and ensure trustworthy results.

1. Clarify experiment goal and context

Ask clarifying questions about the pricing change, target population, and business objectives to ensure the metric hierarchy aligns with strategic priorities.

2. Define primary success metric

Choose a single primary metric that directly measures the experiment's goal, such as revenue per user or conversion rate, and justify why it is the most relevant.

3. Identify guardrail metrics

Select guardrail metrics that capture potential negative side effects, such as customer satisfaction, churn, or long-term retention, and set acceptable thresholds for each.

4. Set decision thresholds and statistical criteria

Determine the minimum detectable effect, significance level (e.g., alpha=0.05), power (e.g., 80%), and practical significance threshold for the primary metric, and specify how guardrails will be monitored.

5. Plan for analysis and decision-making

Outline the analysis approach, including handling multiple comparisons, novelty effects, and segment analysis, and describe the go/no-go decision rule based on the thresholds.

Key Points to Mention

  • Primary metric should be directly tied to the experiment's goal and business objective.
  • Guardrail metrics must include both user experience (e.g., satisfaction, churn) and long-term business health (e.g., retention, LTV).
  • Decision threshold should combine statistical significance (p-value) with practical significance (effect size) and consider the cost of false positives/negatives.
  • Pre-register the analysis plan to avoid p-hacking and ensure validity.
  • Consider multiple testing corrections (e.g., Bonferroni, FDR) when evaluating multiple metrics.
  • Account for novelty effects and seasonality by running the experiment for sufficient duration and using holdout groups.

AI-generated suggestions, not part of the candidate's original notes. May be inaccurate — verify before relying on them.

Q3

Why is a 2-week runtime insufficient for this experiment, and what is the minimum viable schedule you would propose? Include your MDE assumptions, traffic constraints, unit of randomization, bucketing persistence strategy, and how you would handle sequential looks.

A/B Testing & ExperimentationProduct Analytics & Metrics
Author's notes

Two weeks is obviously not enough to observe churn, which is kind of the whole point of a pricing test.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by explaining why 2 weeks is insufficient: low power to detect the MDE, novelty effects, and weekly seasonality not fully captured. Then propose a minimum viable schedule based on power analysis, traffic constraints, and sequential testing considerations, while detailing MDE, unit of randomization, bucketing persistence, and sequential look handling.

Pro tip: Emphasize that the minimum duration should be driven by the number of complete weeks needed to capture seasonality and achieve sufficient power, not just a fixed number of days. Also, mention that sequential looks require alpha-spending adjustments to control false positives.

1. Explain why 2 weeks is insufficient

Discuss statistical power: with typical traffic and MDE, 2 weeks may not yield enough samples to detect the desired effect. Also, 2 weeks may not cover full weekly cycles and can be biased by novelty or primacy effects.

2. Define MDE and power assumptions

State your assumed MDE (e.g., 2% relative lift), significance level (e.g., 5%), and power (e.g., 80%). Use these to compute required sample size per variant.

3. Estimate traffic and required duration

Based on daily eligible traffic and unit of randomization (e.g., user), calculate how many days are needed to reach the required sample size. Round up to full weeks to account for weekly seasonality.

4. Specify unit of randomization and bucketing persistence

Choose a stable unit like user ID or cookie, and describe how bucketing is persisted (e.g., hashing user ID with a salt) to ensure consistent assignment across sessions and devices.

5. Propose sequential testing plan

If sequential looks are needed, use alpha-spending functions (e.g., O'Brien-Fleming) or group sequential boundaries to control Type I error. Otherwise, recommend a fixed-horizon test to avoid peeking.

Key Points to Mention

  • Statistical power and sample size calculation based on MDE, alpha, and power.
  • Weekly seasonality: need at least 2 full weeks, but often 4 weeks to capture monthly patterns.
  • Unit of randomization: typically user-level to avoid interference and ensure consistent experience.
  • Bucketing persistence: deterministic hashing of user ID to maintain assignment across sessions.
  • Sequential testing: alpha-spending or group sequential methods to adjust for multiple looks.
  • Novelty and primacy effects: longer duration helps mitigate these biases.

AI-generated suggestions, not part of the candidate's original notes. May be inaccurate — verify before relying on them.

Q4

Describe the full instrumentation and analysis plan: eligibility and exposure rules, identity resolution to prevent bucket switching, handling of discounts and coupons, seasonality controls, and how you would estimate LTV during the experiment while accounting for right-censoring.

A/B Testing & ExperimentationData ModelingProduct Analytics & Metrics
Author's notes

Right-censoring is the part I found most interesting and also most underspecified in my answer.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Structure your answer around the experiment lifecycle: pre-experiment design, in-experiment monitoring, and post-experiment analysis. Emphasize rigorous identity resolution, seasonality controls, and survival analysis for LTV to show depth in causal inference and business metrics.

Pro tip: Mention that you would pre-register the analysis plan and use sequential testing to avoid peeking, and that you would validate identity resolution by checking for bucket switching rates and using stable user IDs.

1. Define Eligibility and Exposure

Specify who is eligible for the experiment (e.g., new users, specific regions) and how exposure is logged (e.g., first visit after launch). Ensure exposure is triggered only once per user to avoid multiple exposures.

2. Identity Resolution and Bucket Assignment

Use a stable user identifier (e.g., Google account ID) for randomization. Implement checks to prevent bucket switching: log all identifiers, use deterministic hashing, and monitor for inconsistencies across devices/sessions.

3. Handle Discounts, Coupons, and Seasonality

Stratify randomization by discount eligibility or include discount usage as a covariate. Control for seasonality by using a holdback group, time-based fixed effects, or comparing to historical benchmarks.

4. Estimate LTV with Right-Censoring

Use survival analysis (e.g., Kaplan-Meier, Cox model) or parametric models to estimate LTV over a fixed horizon, accounting for users who haven't churned yet. Alternatively, use a shorter proxy metric validated against long-term LTV.

5. Validate and Monitor

Run A/A tests, check sample ratio mismatch, and monitor key metrics daily. Pre-register the analysis plan and use sequential testing to allow early stopping without inflating false positives.

Key Points to Mention

  • Use of stable user IDs and deterministic hashing for bucket assignment to prevent switching.
  • Stratification or covariate adjustment for discounts/coupons to avoid confounding.
  • Seasonality controls: holdback groups, time fixed effects, or historical comparisons.
  • Survival analysis methods (Kaplan-Meier, Cox) for LTV with right-censoring.
  • Pre-registration of analysis plan and sequential testing to avoid p-hacking.
  • Sample ratio mismatch checks and A/A tests for validation.

AI-generated suggestions, not part of the candidate's original notes. May be inaccurate — verify before relying on them.

Q5

If the pricing test results are positive, what does your rollout and escalation plan look like, and how would you monitor for long-run effects or customer backlash after full launch?

A/B Testing & ExperimentationPricing & MonetizationStakeholder Management
Author's notes

I kept this answer shorter than I should have.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by framing the rollout as a phased, data-driven process that balances speed with risk mitigation, then outline specific monitoring metrics and guardrails for long-term effects. Emphasize cross-functional collaboration and a clear escalation path to address any negative signals.

Pro tip: Propose a 'holdback' group that retains the old pricing for a small percentage of users long after launch, allowing you to measure long-term effects and detect backlash without risking the full user base.

1. Phased Rollout Plan

Propose a gradual rollout (e.g., 5%, 25%, 50%, 100%) with predefined success metrics and go/no-go criteria at each stage. Include a timeline and responsible stakeholders.

2. Monitoring and Guardrail Metrics

Define key metrics to track post-launch, such as conversion, revenue, customer satisfaction (CSAT), churn, and support ticket volume. Set thresholds for automatic alerts and pauses.

3. Long-Term Effect Measurement

Design a long-term holdback experiment or cohort analysis to measure effects over months, including seasonality and customer lifetime value (LTV). Use statistical methods to detect gradual changes.

4. Backlash Detection and Response

Monitor social media, reviews, and customer feedback for negative sentiment. Establish a rapid response team and communication plan to address concerns and adjust if needed.

5. Escalation and Decision Framework

Create a clear escalation path: if guardrails are breached, pause rollout, investigate root cause, and decide whether to adjust, revert, or continue with mitigations. Include stakeholders from product, engineering, and marketing.

Key Points to Mention

  • Phased rollout with clear go/no-go criteria at each stage
  • Guardrail metrics (e.g., churn, CSAT, support tickets) with automated alerts
  • Long-term holdback group to measure sustained effects and LTV
  • Sentiment analysis and social listening for early backlash detection
  • Cross-functional escalation matrix and communication plan
  • Statistical methods to distinguish signal from noise in long-term data

AI-generated suggestions, not part of the candidate's original notes. May be inaccurate — verify before relying on them.

Q6

Design a 2x2 factorial experiment for a signup CTA with two factors: button color (red vs. blue) and position (top vs. bottom). Cover hypotheses for main and interaction effects, randomization, power and MDE per effect, multiple testing correction, downstream signup analysis, device stratification, and when you'd prefer chained tests over a full factorial.

A/B Testing & ExperimentationProduct Analytics & MetricsTechnical Trade-offs
Author's notes

Switching gears to this after the pricing monster was a bit of a relief, but the interaction effect question is where people usually slip up.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by framing the experiment as a 2x2 factorial design with four cells, then clearly state hypotheses for main effects (color, position) and interaction. Walk through the design choices—randomization, power/MDE per effect, multiple testing correction—and explain how you'd analyze downstream signup metrics with device stratification. Conclude with when chained tests are preferable to a full factorial.

Pro tip: Emphasize that interaction effects require larger sample sizes and that you should pre-register the analysis plan, including correction method, to avoid p-hacking. Also, mention that device stratification can reveal heterogeneous treatment effects that might be masked in aggregate.

1. Define hypotheses and design

State null and alternative hypotheses for main effects of color and position, and for their interaction. Describe the 2x2 factorial design with four cells: red-top, red-bottom, blue-top, blue-bottom.

2. Randomization and sample size

Explain random assignment of users to one of the four cells, ensuring balanced groups. Discuss power analysis to determine sample size needed to detect main effects and interaction, including MDE per effect.

3. Analysis plan and corrections

Outline the statistical tests (e.g., ANOVA or regression with interaction term) and multiple testing correction (e.g., Bonferroni, Holm-Bonferroni, or FDR) to control family-wise error rate across the three effects.

4. Downstream metrics and stratification

Define primary metric (e.g., signup conversion) and guardrail metrics. Explain how to analyze downstream signup behavior, including device stratification to check for heterogeneous treatment effects and ensure robustness.

5. When to use chained tests

Discuss scenarios where chained tests (sequential testing of one factor at a time) are preferred over full factorial, such as limited traffic, need for faster iteration, or when interaction is unlikely or not of interest.

Key Points to Mention

  • Main effects: color and position; interaction effect: color x position
  • Randomization unit (user-level) and balanced allocation
  • Power analysis: sample size for main effects vs. interaction (interaction requires ~4x sample)
  • Multiple testing correction: Bonferroni, Holm-Bonferroni, or FDR
  • Downstream signup analysis: conversion rate, time to signup, retention
  • Device stratification: mobile vs. desktop, potential heterogeneous effects
  • Chained tests: when traffic is limited, interaction unlikely, or sequential learning desired

AI-generated suggestions, not part of the candidate's original notes. May be inaccurate — verify before relying on them.