← Meta Interview Insights

Meta·Data Scientist·Onsite - Product Sense / Strategy·Senior

SeniorPrefer not to say
May 2026

Summary

A Meta DS onsite that was essentially one giant product metrics case study disguised as six separate questions. The whole thing revolved around a friend-activity notification feature, and they went deep enough that I left genuinely unsure if I'd done well or completely bombed it.

Questions Asked (6)

Q1

You receive a notification that a friend is attending a local event. Define what 'high quality' means for this notification system: specify a North Star metric, three driver metrics, and three guardrail metrics with exact formulas, denominator definitions, and attribution windows. Also distinguish leading from lagging indicators.

Product Analytics & MetricsProduct Sense & Ideation
Author's notes

This is where I spent the most time and probably still left gaps.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by clarifying the notification's purpose and the user action it aims to drive, then define a North Star metric that captures the core value. Structure your answer by proposing one North Star, three driver metrics that directly influence it, and three guardrail metrics that ensure no harm, specifying exact formulas, denominators, and attribution windows. Finally, classify each metric as leading or lagging to show strategic thinking.

Pro tip: Tie your metrics to Meta's emphasis on meaningful social interactions and long-term user value; avoid vanity metrics like raw notification opens and instead focus on downstream engagement like event attendance or re-engagement with the friend.

1. Clarify the goal and user journey

Define the notification's objective (e.g., drive event attendance or strengthen social ties) and map the user journey from receiving the notification to taking action. This ensures metrics align with business and user value.

2. Define the North Star metric

Choose a single metric that best captures the notification's success, such as the rate of users who attend the event after receiving the notification. Provide an exact formula, denominator, and attribution window.

3. Identify driver metrics

Select three metrics that directly influence the North Star, such as notification open rate, event page visit rate, and RSVP rate. For each, specify formula, denominator, and attribution window.

4. Identify guardrail metrics

Choose three metrics to monitor for negative side effects, such as notification dismissal rate, user report rate, and notification opt-out rate. Define each with formula, denominator, and attribution window.

5. Classify leading vs. lagging indicators

Explain which metrics are leading (predictive of future success, e.g., open rate) and which are lagging (outcome-based, e.g., event attendance). This shows understanding of metric timing and causality.

Key Points to Mention

  • North Star metric should reflect meaningful social interaction, e.g., 'Event Attendance Rate from Friend Notification' = (number of users who attend the event within 7 days of receiving the notification) / (number of users who received the notification).
  • Driver metrics: Notification Open Rate (opens / delivered, 24-hour window), Event Page Visit Rate (visits / opens, 24-hour window), RSVP Rate (RSVPs / visits, 48-hour window).
  • Guardrail metrics: Notification Dismissal Rate (dismissals / delivered, 24-hour window), User Report Rate (reports / delivered, 7-day window), Notification Opt-Out Rate (opt-outs / delivered, 30-day window).
  • Leading indicators: open rate, click-through rate (predict future attendance). Lagging indicators: event attendance, RSVP rate (outcome measures).
  • Attribution windows should be justified based on user behavior; e.g., 7 days for attendance to allow planning, 24 hours for opens to capture immediate engagement.
  • Consider counterfactual or holdout groups to measure incremental impact, as correlation does not imply causation.

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

Q2

Before building this feature, estimate adoption and size the opportunity using historical event data. The naive estimator averages the percentage of sign-ups who arrived with at least one friend across past events. What are at least four biases in that estimator, and how would you correct for them? Also show how you'd compute a 95% confidence interval for expected daily notification opens.

Product Analytics & MetricsA/B Testing & Experimentation
Author's notes

Four biases came to me pretty quickly: event-type confounding (concerts vs.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

First, identify and explain the biases in the naive estimator, such as selection bias, survivorship bias, and confounding, and propose corrections like stratification, weighting, or causal inference methods. Then, for the confidence interval, outline the appropriate statistical method (e.g., normal approximation or bootstrap) and the necessary assumptions, ensuring to mention data requirements and potential pitfalls.

Pro tip: Show that you understand the business context by linking each bias to a specific product decision and quantifying the potential impact of the correction. Also, when computing the confidence interval, clarify whether you're treating daily opens as independent or if there's temporal correlation, and adjust accordingly.

1. Identify biases in the naive estimator

List at least four biases: selection bias (non-random event participation), survivorship bias (only successful events included), confounding (event size or seasonality), and measurement bias (inaccurate friend counts). Explain how each skews the estimate.

2. Propose corrections for each bias

For each bias, suggest a correction: e.g., use stratified sampling or inverse probability weighting for selection bias; include all events or model event success for survivorship; control for confounders via regression or matching; validate friend data for measurement bias.

3. Estimate expected daily notification opens

Define the metric: average number of notification opens per day per user. Decide on the data source and time window, and consider whether to model at user-level or aggregate level.

4. Compute 95% confidence interval

Choose a method: if opens are approximately normal, use mean ± 1.96*SE; if skewed or count data, use Poisson or negative binomial model; if temporal correlation, use time series or bootstrap. State assumptions and show calculation.

5. Validate and communicate uncertainty

Check assumptions (e.g., normality, independence) and consider sensitivity analysis. Communicate the CI in business terms, noting that it reflects uncertainty in the estimate.

Key Points to Mention

  • Selection bias: events with more friend sign-ups may attract different users; correct with weighting or stratification.
  • Survivorship bias: only successful events in data; correct by including failed events or modeling event success.
  • Confounding: event size, seasonality, or marketing spend affect friend sign-ups; correct via regression adjustment or matching.
  • Measurement bias: friend counts may be inaccurate; correct by validating data or using alternative metrics.
  • For CI: use normal approximation if n large and variance known; otherwise use bootstrap or Poisson model.
  • Consider temporal correlation in daily opens; use time series methods or cluster-robust standard errors.

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

Q3

Justify the engineering expense for building this feature. Derive a break-even formula that links expected incremental opens, sessions, and retention to dev-weeks and infrastructure cost, and state the minimum detectable effect needed to proceed.

Product StrategyProduct Analytics & Metrics
Author's notes

I sketched out a rough formula: incremental retention lift times LTV per retained user, summed over the expected user base, has to exceed dev cost plus infra cost over some payback period.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by framing the feature's value in terms of incremental business metrics (opens, sessions, retention) and then translate those into a break-even condition where expected incremental revenue equals engineering and infrastructure costs. Derive a formula that solves for the minimum detectable effect (MDE) required to justify the investment, and discuss how to measure it with an experiment.

Pro tip: Anchor your answer in the company's north-star metric and show that you understand the trade-off between statistical power and development cost—this demonstrates product sense and analytical rigor.

1. Define the incremental value model

Express the feature's expected impact as incremental opens, sessions, and retention, and map these to a monetary value (e.g., via ad revenue per session or LTV).

2. Quantify costs

List all costs: dev-weeks (converted to dollars using fully-loaded engineer cost) and ongoing infrastructure cost (e.g., cloud, storage, compute).

3. Derive break-even formula

Set total incremental value equal to total cost and solve for the required effect size. For example: (Δopens * V_open + Δsessions * V_session + Δretention * V_retention) * N = C_dev + C_infra.

4. Solve for minimum detectable effect (MDE)

Rearrange the formula to isolate the MDE (e.g., Δsessions) given baseline metrics, costs, and expected user base. This MDE is the threshold the experiment must detect to justify proceeding.

5. Validate with experiment design

Check if the MDE is achievable given traffic and variance; if not, recommend not proceeding or re-scoping the feature.

Key Points to Mention

  • Incremental metrics: focus on opens, sessions, and retention as key drivers of engagement and monetization.
  • Cost components: include fully-loaded dev-weeks (salary, benefits, overhead) and infrastructure (servers, storage, bandwidth).
  • Break-even condition: total incremental value = total cost; solve for the required effect size.
  • Minimum detectable effect (MDE): the smallest effect the experiment can reliably detect given sample size and power.
  • Statistical power and significance: ensure the MDE is detectable with 80% power and 95% confidence to avoid false negatives.
  • Opportunity cost: consider alternative uses of engineering resources and the expected ROI of this feature versus others.

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

Q4

Describe a pre-launch evidence plan using concept tests, surveys, and analog notification benchmarks to predict CTR uplift. How would you adjust for response bias and population mismatch?

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

I talked about using survey intent data discounted by a response bias correction factor, basically applying a historical ratio of stated-to-actual behavior from prior similar features.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by outlining a structured pre-launch evidence plan that combines concept tests, surveys, and analog benchmarks to estimate CTR uplift. Then, discuss how to adjust for response bias and population mismatch using statistical techniques and validation steps. Emphasize the iterative nature of refining predictions as more data becomes available.

Pro tip: Leverage Meta's internal tools and historical data to calibrate analog benchmarks, and always validate predictions with a small-scale A/B test before full launch to mitigate risks.

1. Define the Evidence Plan

Outline the components: concept tests to gauge initial appeal, surveys to measure intent, and analog benchmarks from similar past launches to predict CTR uplift.

2. Collect and Analyze Data

Execute concept tests and surveys, ensuring representative sampling. Gather analog benchmarks from historical data, adjusting for differences in context.

3. Adjust for Biases

Identify potential response bias (e.g., social desirability) and population mismatch (e.g., test group vs. target). Apply weighting, stratification, or modeling techniques to correct.

4. Predict and Validate

Combine adjusted data to predict CTR uplift. Validate predictions with a holdout or small-scale A/B test to assess accuracy and refine the model.

5. Iterate and Communicate

Refine the plan based on validation results. Communicate assumptions, limitations, and confidence intervals to stakeholders.

Key Points to Mention

  • Use of concept tests (e.g., A/B tests of ad creatives) to measure early signals of CTR uplift.
  • Surveys to capture user intent and preferences, with attention to sampling methods to reduce bias.
  • Analog benchmarks from similar products or campaigns, adjusting for temporal and contextual differences.
  • Techniques to adjust for response bias: weighting, propensity score adjustment, or using control questions.
  • Methods to address population mismatch: re-weighting, stratification, or using synthetic control groups.
  • Importance of validating predictions with a small-scale A/B test before full launch.

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

Q5

Design the A/B test for this notification feature: define the randomization unit, primary and guardrail metrics, sample sizing, and explain when cluster randomization is required due to network effects. How would you detect interference and what do you do if contamination is found?

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

Network effects question at Meta, predictable but still tricky to get right.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by clarifying the notification feature and its potential for network effects, then systematically define the randomization unit, metrics, sample size, and interference detection plan. Emphasize the trade-offs between user-level and cluster randomization, and outline a robust process for detecting and mitigating contamination.

Pro tip: Always consider the social graph at Meta: even if the feature seems individual, interactions like notifications can spill over. Proactively discuss cluster randomization and interference detection to show depth.

1. Clarify feature and network effects

Understand the notification feature's mechanics and how it might create interference between users (e.g., sharing, social pressure). Determine if network effects are likely.

2. Choose randomization unit

Decide between user-level and cluster randomization based on interference risk. If interference is expected, use cluster randomization (e.g., by social community or geographic region).

3. Define metrics and sample size

Select primary metric (e.g., notification engagement) and guardrail metrics (e.g., user retention, notification fatigue). Calculate sample size using power analysis, accounting for cluster design if applicable.

4. Detect interference

Plan methods to detect interference: compare outcomes between treated and control clusters, use social network analysis, or run A/A tests with cluster randomization. Monitor for spillover effects.

5. Mitigate contamination

If contamination is found, consider re-randomizing, using cluster-based analysis, or adjusting the experiment design (e.g., switch to cluster randomization). Document and learn from the issue.

Key Points to Mention

  • Randomization unit: user vs. cluster, and when cluster is needed due to network effects
  • Primary metric: e.g., click-through rate on notifications; guardrail metrics: e.g., unsubscribe rate, user satisfaction
  • Sample sizing: power analysis, minimum detectable effect, intra-cluster correlation for cluster randomization
  • Interference detection: A/A tests, social network analysis, spillover measurement
  • Contamination mitigation: re-randomization, cluster-level analysis, design adjustments
  • Meta-specific considerations: social graph, News Feed, notification delivery systems

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

Q6

If an experiment shows time-on-site increases but CTR on the notification decreases, how do you make a go/no-go decision? Walk through a metric hierarchy or utility function, including constraints on unsubscribe rates and notification volume.

Product Analytics & MetricsA/B Testing & ExperimentationAdaptability & Ambiguity
Author's notes

Blanked for a second here.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by clarifying the experiment's goal and defining a clear metric hierarchy with a primary success metric (e.g., long-term user engagement) and guardrail metrics (unsubscribe rate, notification volume). Then, evaluate the trade-offs using a utility function that weights the primary metric against guardrails, and make a go/no-go decision based on whether the net utility is positive and guardrails are not violated.

Pro tip: Emphasize that short-term CTR decreases might be acceptable if long-term engagement increases, but always check for novelty effects and ensure guardrails are not breached. Consider segment-level analysis to see if the trade-off varies across user groups.

1. Clarify experiment goal and define metric hierarchy

Identify the primary objective (e.g., increase long-term user engagement) and establish a hierarchy: primary metric (time-on-site), secondary metrics (CTR), and guardrail metrics (unsubscribe rate, notification volume).

2. Assess guardrail metrics and constraints

Check if unsubscribe rate and notification volume are within acceptable thresholds. If guardrails are violated, the experiment may be a no-go regardless of primary metric improvement.

3. Construct a utility function to weigh trade-offs

Define a utility function that combines the primary metric and secondary metrics, with weights reflecting business priorities. For example, utility = w1 * Δtime-on-site - w2 * ΔCTR - w3 * Δunsubscribe rate.

4. Evaluate statistical significance and practical significance

Ensure the observed changes are statistically significant and consider the effect sizes. A small but significant increase in time-on-site might not outweigh a large decrease in CTR.

5. Make a go/no-go decision and consider next steps

If net utility is positive and guardrails are satisfied, recommend a go; otherwise, no-go. Suggest further analysis (e.g., segment-level, long-term holdout) if ambiguous.

Key Points to Mention

  • Define a clear primary metric (e.g., long-term user engagement) and guardrail metrics (unsubscribe rate, notification volume).
  • Use a utility function to quantify trade-offs, with weights based on business priorities.
  • Check guardrail metrics first; if violated, consider no-go even if primary metric improves.
  • Consider novelty effects and long-term impact; short-term CTR decrease might be acceptable if engagement increases.
  • Perform segment-level analysis to identify heterogeneous treatment effects.
  • Recommend a long-term holdout or follow-up experiment if results are ambiguous.

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