← Gusto Interview Insights

Gusto·Data Scientist·Technical Phone Screen·Senior

Senior
May 2026

Summary

Gusto data scientist interview with a meaty A/B testing question that covered the full stack: stats, multiple comparisons, confidence intervals, and a ship decision with stakeholder communication. Pretty involved for a single problem.

Questions Asked (4)

Q1

Given aggregated results from an A/B/n experiment with one control and two treatment variants (each with 50k users and differing conversion counts), compute p-values for each variant versus control using a proportions test in Python.

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

This part I felt okay about.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by clarifying the data structure and the null hypothesis (no difference between each treatment and control). Then, for each variant, compute the z-statistic for the difference in proportions and derive the two-sided p-value using the normal approximation. Finally, discuss multiple testing correction and practical significance.

Pro tip: Always apply a multiple comparisons correction (e.g., Bonferroni or Benjamini-Hochberg) when testing multiple variants against a single control, and report both raw and adjusted p-values to avoid false positives.

1. Clarify the data and hypotheses

Confirm the conversion counts and sample sizes for control and each treatment. State the null hypothesis: the conversion rate for each treatment equals that of the control.

2. Compute conversion rates and standard errors

Calculate the conversion rate for each group (conversions / 50,000). Compute the pooled proportion and standard error for each treatment-control comparison.

3. Calculate z-statistics and p-values

For each variant, compute the z-statistic as (p_treatment - p_control) / SE, then find the two-sided p-value using the normal CDF.

4. Adjust for multiple comparisons

Apply a correction method (e.g., Bonferroni or Benjamini-Hochberg) to control the family-wise error rate or false discovery rate across the two comparisons.

5. Interpret results and discuss practical significance

Report p-values, confidence intervals, and effect sizes. Consider whether the observed differences are practically meaningful, not just statistically significant.

Key Points to Mention

  • Two-proportion z-test and its assumptions (independent samples, large sample size).
  • Pooled vs. unpooled variance estimation for the standard error.
  • Multiple testing correction (Bonferroni, Holm, or Benjamini-Hochberg).
  • Confidence intervals for the difference in proportions.
  • Practical significance vs. statistical significance (effect size, business impact).
  • Python implementation using scipy.stats or statsmodels (e.g., proportions_ztest).

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

Q2

With two variants being tested simultaneously, how do you handle the multiple comparisons problem? Apply a correction method and report your adjusted conclusions.

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

Bonferroni is the easy answer and I went with it, adjusted alpha to 0.025 per test.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by explaining the multiple comparisons problem in the context of A/B testing with two variants, then propose a correction method like Bonferroni or Benjamini-Hochberg, and finally demonstrate how to apply it to adjust p-values and draw conclusions. Emphasize the trade-off between controlling false positives and maintaining statistical power.

Pro tip: Mention that the choice of correction depends on the goal: Bonferroni is conservative and best when false positives are costly, while FDR methods like Benjamini-Hochberg are more powerful when exploring multiple metrics. Also, consider using sequential testing or Bayesian methods as alternatives.

1. Identify the multiple comparisons issue

Explain that testing two variants simultaneously increases the chance of false positives due to multiple hypotheses (e.g., each variant vs. control). Quantify the inflated Type I error rate if uncorrected.

2. Choose an appropriate correction method

Select a method based on the context: Bonferroni for strict family-wise error control, or Benjamini-Hochberg for false discovery rate control when multiple metrics are tested. Justify your choice.

3. Apply the correction to p-values

Show how to adjust the p-values: for Bonferroni, multiply each p-value by the number of comparisons (or divide alpha); for Benjamini-Hochberg, rank p-values and compare to (i/m)*Q. Provide a concrete example.

4. Interpret adjusted results and draw conclusions

Compare adjusted p-values to the significance level and state which variants are statistically significant. Discuss the impact on conclusions, including potential loss of power and the need for larger sample sizes.

5. Discuss trade-offs and alternatives

Acknowledge the trade-off between Type I and Type II errors. Mention alternatives like sequential testing, Bayesian methods, or pre-registering a primary metric to avoid the issue.

Key Points to Mention

  • Family-wise error rate (FWER) and false discovery rate (FDR) concepts
  • Bonferroni correction: formula and when to use it
  • Benjamini-Hochberg procedure: steps and when it's preferred
  • Impact on statistical power and sample size requirements
  • Pre-registration of hypotheses and primary metrics to mitigate multiple comparisons
  • Alternatives: sequential testing, Bayesian A/B testing, or hierarchical models

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

Q3

Construct 95% confidence intervals for the absolute and relative uplift of each variant versus control.

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

Blanked for a second on the relative uplift CI formula.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by clarifying the metric and data structure, then compute the absolute uplift (difference in means) and relative uplift (ratio of means) for each variant vs control. Use the delta method or bootstrap to derive standard errors and construct 95% confidence intervals, ensuring to account for multiple comparisons if needed.

Pro tip: Always check the assumptions behind the delta method (e.g., large sample, non-zero control mean) and consider using bootstrap for robustness, especially with skewed metrics. Also, present both absolute and relative intervals because stakeholders often care about different aspects.

1. Clarify metric and data

Confirm the metric (e.g., conversion rate, revenue per user) and the experimental design (sample sizes, randomization). Ensure you have the necessary summary statistics (means, variances, sample sizes) for each group.

2. Compute absolute uplift and its CI

Calculate the difference in means between variant and control. For the CI, use the standard error of the difference: sqrt(var_variant/n_variant + var_control/n_control), then construct a 95% CI using the normal approximation (or t-distribution if small samples).

3. Compute relative uplift and its CI

Calculate the ratio of means (variant/control). Use the delta method to approximate the variance of the ratio: (mean_variant^2 / mean_control^2) * (var_variant/(n_variant*mean_variant^2) + var_control/(n_control*mean_control^2)). Alternatively, use bootstrap resampling to empirically derive the CI.

4. Adjust for multiple comparisons if needed

If comparing multiple variants to a single control, consider using a correction like Bonferroni or Dunnett's test to control the family-wise error rate. This affects the confidence level or the critical value used.

5. Interpret and communicate results

Present the CIs for both absolute and relative uplift, noting whether they include zero (for absolute) or one (for relative). Discuss practical significance and any caveats (e.g., assumptions, sample size).

Key Points to Mention

  • Difference in means for absolute uplift and ratio of means for relative uplift
  • Standard error calculation for difference and delta method for ratio
  • Bootstrap as an alternative for non-normal metrics or small samples
  • Assumptions: independence, large sample, non-zero control mean
  • Multiple comparison corrections (e.g., Bonferroni, Dunnett)
  • Interpretation: CI for absolute includes 0 vs relative includes 1

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

Q4

Make a ship or no-ship recommendation based on the experiment results, and walk through what additional checks you'd run before actually launching. How would you communicate uncertainty to non-technical stakeholders?

A/B Testing & ExperimentationStakeholder ManagementProduct Analytics & Metrics
Author's notes

This is where the conversation got interesting.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by clearly stating your ship/no-ship recommendation based on the primary metric and guardrails, then outline the additional checks you'd run to validate the result (e.g., novelty effects, segment analysis, SRM). Finally, explain how you'd communicate uncertainty to non-technical stakeholders using plain language, confidence intervals, and business impact framing.

Pro tip: Always tie the recommendation to business impact and risk tolerance—stakeholders care more about potential revenue or user experience changes than p-values. Use a 'confidence score' or 'range of outcomes' to make uncertainty tangible.

1. State the recommendation

Give a clear ship or no-ship call based on the primary metric and guardrail metrics. If results are ambiguous, recommend a conditional ship (e.g., ship to a subset) or extend the test.

2. Validate the experiment

Run additional checks: sample ratio mismatch (SRM), novelty/primacy effects, segment-level consistency, and sensitivity analyses. Ensure the result is robust and not driven by outliers or a single segment.

3. Quantify uncertainty

Calculate confidence intervals and effect sizes. Consider Bayesian methods for a more intuitive probability of superiority. Assess practical significance vs. statistical significance.

4. Communicate to stakeholders

Translate statistical uncertainty into business terms: e.g., 'We are 85% confident the change will increase conversion by 1-3%, which could mean $X in annual revenue.' Use visualizations like confidence intervals or probability distributions.

5. Propose next steps

Suggest a phased rollout with monitoring, or additional tests to reduce uncertainty. Define success metrics and guardrails for the rollout phase.

Key Points to Mention

  • Primary metric and guardrail metrics (e.g., conversion, revenue, churn, latency)
  • Statistical significance vs. practical significance
  • Sample ratio mismatch (SRM) and other validity checks
  • Novelty effects and segment analysis (e.g., new vs. existing users)
  • Confidence intervals and Bayesian probability
  • Business impact framing and risk tolerance

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