← Pinterest Interview Insights

Pinterest·Data Scientist·Technical Phone Screen·Senior

SeniorPrefer not to say
Jun 2026Remote

Summary

Pinterest DS interview focused almost entirely on a single dense A/B testing case study about video pins in the home feed. Five sub-questions stacked on top of each other, covering everything from hypothesis setup to SRM checks to multiple comparison corrections. Felt more like a take-home problem being walked through live than a conversational interview.

Questions Asked (5)

Q1

State the null and alternative hypotheses for the primary metric precisely, including the direction of the test and the minimum detectable effect.

A/B Testing & Experimentation
Author's notes

I fumbled this a little by writing H1 as just 'Treatment > Control' without anchoring it to the MDE.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by clearly defining the primary metric and its current baseline value. Then state the null hypothesis (no effect or non-inferiority) and the alternative hypothesis (two-sided or one-sided) based on the business goal. Finally, specify the minimum detectable effect (MDE) in absolute and relative terms, and justify it using practical significance and power analysis.

Pro tip: Always tie the MDE to a business-relevant threshold (e.g., revenue impact or user engagement lift) rather than just statistical convention, and mention that the direction of the test should align with the product team's expectation to avoid misleading conclusions.

1. Define the primary metric

Identify the key metric (e.g., click-through rate, saves per user) and its current baseline value from historical data. Ensure it's a single, well-defined metric that directly measures the experiment's goal.

2. State the null hypothesis (H0)

Formulate H0 as no difference or no effect between control and treatment groups. For a two-sided test, H0: μ_treatment = μ_control; for a one-sided test, specify the direction (e.g., H0: μ_treatment ≤ μ_control).

3. State the alternative hypothesis (H1)

Formulate H1 as the presence of an effect. For a two-sided test, H1: μ_treatment ≠ μ_control; for a one-sided test, H1: μ_treatment > μ_control (or <). The direction should be based on the expected impact.

4. Specify the minimum detectable effect (MDE)

Determine the smallest effect size that is practically meaningful and that the experiment is powered to detect. Express MDE in absolute terms (e.g., 0.5% increase) and relative terms (e.g., 5% lift), and justify it with business impact or prior data.

5. Justify the direction and MDE

Explain why the test is one-sided or two-sided, and how the MDE was chosen (e.g., based on power analysis, cost of false positives/negatives, or stakeholder requirements). Mention the significance level (α) and power (1-β) used.

Key Points to Mention

  • Primary metric definition and baseline value
  • Null and alternative hypotheses with correct notation
  • Direction of the test (one-sided vs. two-sided) and rationale
  • Minimum detectable effect (MDE) in absolute and relative terms
  • Justification for MDE using business relevance or power analysis
  • Significance level (α) and statistical power (1-β)

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

Q2

Run a chi-square test to check for sample ratio mismatch given the observed assignment counts, and explain whether SRM should be suspected.

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

This part went okay.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

First, state the null hypothesis that the observed assignment counts follow the expected ratio. Then, compute the chi-square statistic and p-value, and compare the p-value to a pre-defined significance level (e.g., 0.05) to decide whether to reject the null. Finally, interpret the result in the context of SRM: if p-value is below threshold, suspect SRM and investigate potential causes.

Pro tip: Always check the expected counts and ensure they are sufficiently large (e.g., >5) for the chi-square test to be valid. Also, mention that SRM detection should be done early and often, and that even a small deviation can be practically significant in large-scale experiments.

1. State the hypothesis

Define the null hypothesis (H0): the observed counts match the expected ratio (e.g., 50/50). The alternative hypothesis (H1): the observed counts do not match the expected ratio.

2. Calculate expected counts

Compute the expected number of users in each group based on the total sample size and the intended allocation ratio.

3. Compute chi-square statistic

Use the formula: χ² = Σ (observed - expected)² / expected. Sum over all groups.

4. Determine p-value and compare

Find the p-value from the chi-square distribution with degrees of freedom = number of groups - 1. Compare to significance level (e.g., α=0.05).

5. Interpret and decide

If p-value < α, reject H0 and suspect SRM; otherwise, do not suspect SRM. Discuss potential causes and next steps.

Key Points to Mention

  • Null hypothesis: observed counts follow expected ratio
  • Chi-square test assumptions: independent observations, expected counts >5
  • Calculation of chi-square statistic and degrees of freedom
  • P-value interpretation and significance level (e.g., 0.05)
  • SRM implications: data quality issue, invalidates experiment results
  • Practical significance vs statistical significance in large samples

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

Q3

For the primary metric, compute the absolute lift, relative lift, a 95% confidence interval for the difference in means, and the one-sided p-value. Is it significant at alpha = 0.05?

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

Absolute lift is 0.36 minutes, relative lift is 3%, which clears the 2% MDE.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by clearly defining the primary metric and the null and alternative hypotheses. Then compute the absolute lift (difference in means), relative lift (percentage change), standard error, and the 95% confidence interval for the difference. Finally, calculate the one-sided p-value and compare it to alpha = 0.05 to determine significance.

Pro tip: Always state your assumptions (e.g., normality, independence) and mention that for a one-sided test, the p-value is half of the two-sided p-value if the effect is in the hypothesized direction. Also, relate the result to practical significance, not just statistical significance.

1. Define metric and hypotheses

Identify the primary metric (e.g., mean clicks per user) and state the null (H0: μ_treatment - μ_control ≤ 0) and alternative (H1: μ_treatment - μ_control > 0) hypotheses for a one-sided test.

2. Compute absolute and relative lift

Calculate absolute lift as the difference in sample means (treatment - control). Compute relative lift as (absolute lift / control mean) * 100%.

3. Calculate standard error and confidence interval

Compute the standard error of the difference using sample variances and sizes. Construct a 95% confidence interval for the difference: (absolute lift ± t*SE) or (absolute lift ± z*SE) for large samples.

4. Compute one-sided p-value

Calculate the test statistic (t or z) as absolute lift divided by standard error. Find the one-sided p-value as the probability of observing a test statistic at least as extreme in the direction of the alternative.

5. Determine significance

Compare the one-sided p-value to alpha = 0.05. If p < 0.05, reject the null hypothesis and conclude the effect is statistically significant.

Key Points to Mention

  • Difference in means and its standard error formula
  • One-sided vs two-sided tests and when to use each
  • Confidence interval interpretation and relation to hypothesis testing
  • Assumptions: independence, normality, equal variances (or Welch's t-test)
  • Practical significance vs statistical significance
  • Potential pitfalls: multiple testing, peeking, sample ratio mismatch

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

Q4

Run two-proportion z-tests for CTR and D7 retention as secondary metrics, then apply Holm-Bonferroni correction at a familywise alpha of 0.05. Which metrics remain significant after correction?

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

CTR comparison is aggregate clicks over impressions, not per-user, so you use a proportion test on those pooled numbers.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

First, clarify the setup: two metrics (CTR and D7 retention) tested as secondary metrics, each with a two-proportion z-test. Then explain Holm-Bonferroni: order p-values from smallest to largest, compare the smallest to alpha/2, and if significant, compare the next to alpha/1. Finally, state which metrics remain significant based on the corrected thresholds, emphasizing that without actual p-values, the answer is conditional.

Pro tip: Mention that Holm-Bonferroni is uniformly more powerful than Bonferroni and controls the familywise error rate without assuming independence, which is important for correlated metrics like CTR and retention. Also, note that if the smallest p-value exceeds alpha/2, no metrics are significant.

1. Clarify the testing setup

Confirm that CTR and D7 retention are secondary metrics, each analyzed with a two-proportion z-test, and that the familywise alpha is 0.05.

2. Explain Holm-Bonferroni procedure

Describe the step-down method: rank p-values ascending, compare the smallest to alpha/2, the next to alpha/1, and stop at the first non-significant result.

3. Apply to the two metrics

With two metrics, the thresholds are alpha/2 = 0.025 for the smaller p-value and alpha/1 = 0.05 for the larger p-value, provided the smaller is significant.

4. Determine significance

State that the metric with the smaller p-value is significant if p < 0.025; the other is significant only if the first is significant and its p < 0.05.

5. Conclude conditionally

Since actual p-values are not provided, conclude that the answer depends on the observed p-values and summarize the decision rule.

Key Points to Mention

  • Two-proportion z-test assumptions: independent samples, large enough sample size for normal approximation.
  • Holm-Bonferroni controls familywise error rate (FWER) and is more powerful than Bonferroni.
  • With two metrics, the adjusted thresholds are 0.025 and 0.05 (if the first is significant).
  • If the smallest p-value > 0.025, no metrics are significant after correction.
  • If the smallest p-value < 0.025, the second metric is significant only if its p-value < 0.05.
  • Emphasize that without actual p-values, the answer is conditional and requires the data.

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

Q5

Give a ship or no-ship recommendation. How do SRM, novelty effects, and other validity concerns affect your decision, and what additional checks would you run before launching?

A/B Testing & ExperimentationProduct StrategyRoot Cause Analysis
Author's notes

My answer was no-ship pending investigation, mostly because the SRM is a real problem and you can't trust the primary metric result until you know why assignment broke.

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, then systematically address validity threats like SRM, novelty effects, and other biases. Explain how each concern affects your confidence and what additional checks or analyses you would run to mitigate them before making a final call.

Pro tip: Always quantify the impact of validity concerns—e.g., estimate how much of the observed effect could be due to novelty—and propose concrete follow-up experiments or holdout groups to validate long-term effects.

1. State recommendation and primary metric

Begin with a clear ship/no-ship recommendation based on the primary success metric (e.g., CTR, engagement) and its statistical significance. Briefly mention the observed effect size and confidence interval.

2. Check for Sample Ratio Mismatch (SRM)

Verify that the actual traffic split matches the intended ratio using a chi-squared test. If SRM is present, investigate potential causes (e.g., logging issues, bot traffic) and consider invalidating the experiment.

3. Assess novelty and primacy effects

Analyze metric trends over time to detect novelty (initial spike then decay) or primacy (initial dip then recovery) effects. Use cohort analysis or compare early vs. late periods to see if the effect persists.

4. Evaluate other validity threats

Consider other biases such as instrumentation issues, selection bias, or external factors (e.g., seasonality, concurrent experiments). Check for consistency across segments and guardrail metrics.

5. Propose additional checks and final decision

Suggest additional checks like holdout groups, long-term follow-up, or qualitative user research. Weigh the evidence and provide a final recommendation, possibly conditional on further validation.

Key Points to Mention

  • Sample Ratio Mismatch (SRM) detection and its implications for experiment validity
  • Novelty effects: how to identify and adjust for them using time-series analysis
  • Primacy effects and their impact on early experiment results
  • Guardrail metrics and segment-level analysis to ensure no negative impact
  • Statistical power and minimum detectable effect (MDE) considerations
  • Long-term holdout or follow-up experiments to validate sustained impact

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