Meta Data Analyst Interview Questions (2026)

1. 1Recruiter screen30 minEasy

   Process overview, level calibration, and prep guidance — Meta recruiters actively coach on round formats.

2. 2Technical screen45 minHard

   Two DSA problems in 45 minutes on a plain shared editor with no autocomplete or execution.

3. 3Coding round ('Ninja')45 minHard

   Two more problems at loop difficulty; clean near-compilable code and verbalized complexity analysis expected.

4. 4System design ('Pirate')45 minHard

   Design a Meta-scale product system (feed, Stories, chat) with emphasis on read-heavy fan-out, caching, and data modeling.

5. 5Behavioral ('Jedi')45 minMedium

   Deep past-experience discussion on conflict, growth, and impact aligned to Meta values; graded as a real signal round.

1. Q1

   Design an A/B test for a new Facebook Feed ranking or recommendation change. Define hypothesis, primary metric, guardrails, randomization unit, and launch decision rule

   MediumStatistics & Experimentation RoundA/B TestingMeta-specific

   Context: Context: Meta wants to balance engagement, creator health, and advertiser value.

   How to answer: A strong answer will define a clear, testable hypothesis for the Feed ranking change, such as 'The new ranking algorithm will increase user engagement (e.g., time spent, interactions) without negatively impacting user retention or satisfaction.' The primary metric should directly reflect the hypothesis, like 'average daily time spent in Feed per user.' Guardrail metrics are crucial for identifying negative side effects, including 'daily active users (DAU),' 'negative feedback rate,' and 'uninstalls.' The randomization unit should be 'user-ID' to ensure consistent experience, and the launch decision rule will involve statistical significance on the primary metric while ensuring no significant negative movement on guardrail metrics.

2. Q2

   For Instagram Reels, should randomization happen at user, session, device, advertiser, or country level? Explain the tradeoffs

   MediumStatistics & Experimentation RoundA/B TestingMeta-specific

   Context: Consider cross-device behavior, interference, marketplace effects, and operational feasibility.

   How to answer: Randomization for Instagram Reels should primarily happen at the user level to ensure independent observations and avoid contamination. Session or device level randomization can be considered for very short-term, within-session effects but risks user-level contamination if a user has multiple sessions/devices. Advertiser or country level randomization is typically reserved for experiments with network effects or legal/policy changes, as it significantly reduces statistical power and increases the risk of confounding factors. The choice depends on the experiment's goal, the potential for spillover effects, and the desired unit of analysis for metrics.

3. Q3

   Choose primary and guardrail metrics for a WhatsApp Business experiment aimed at improving meaningful social interaction rate. What metrics would prevent a harmful launch?

   MediumStatistics & Experimentation RoundA/B TestingMeta-specific

   Context: Include user experience, partner health, revenue, reliability, and long-term retention considerations.

   How to answer: A strong answer would identify 'Meaningful Social Interaction Rate' (MSI Rate) as the primary metric, defining it as user-initiated, non-automated, two-way communication. Guardrail metrics should include 'Daily Active Users' (DAU), 'Messages Sent/Received per User', 'Block Rate', and 'Report Rate' to ensure the experiment doesn't degrade overall engagement or user safety. The candidate should explain how these guardrails prevent negative impacts like spam or user churn, ensuring a holistic evaluation beyond just the primary metric.

4. Q4

   During a Marketplace experiment, the treatment/control split is 52/48 instead of 50/50. How would you diagnose sample ratio mismatch?

   MediumStatistics & Experimentation RoundA/B TestingMeta-specific

   Context: Assume assignment logs, exposure logs, and eligibility filters may disagree.

   How to answer: To diagnose Sample Ratio Mismatch (SRM), I would first verify the data source and extraction logic to ensure no errors occurred. Next, I'd check for consistent assignment across various dimensions like user ID, device type, and geography, looking for any skewed segments. I would then perform a chi-squared test on the observed counts against the expected 50/50 split to statistically confirm the mismatch. Finally, I'd investigate potential causes such as implementation bugs in the assignment logic, caching issues, or external factors affecting user exposure.

5. Q5

   The Stories experiment is trending positive after two days. A PM wants to stop early and launch. How do you handle peeking and sequential testing?

   MediumStatistics & Experimentation RoundA/B TestingMeta-specific

   Context: Discuss pre-specified stopping rules, alpha spending, business urgency, and risk.

   How to answer: A strong candidate would first acknowledge the PM's enthusiasm but immediately flag the dangers of peeking (increased Type I error rate). They would explain that early stopping invalidates pre-set significance levels and requires adjustments. They should then propose solutions like pre-determining stopping rules (e.g., fixed duration, MDE reached with sufficient power, or using sequential testing methods like A's or SPRT) and emphasize the importance of statistical rigor to ensure reliable results and avoid launching a false positive.

6. Q6

   A new Ads Manager feature shows a large week-1 lift in meaningful social interaction rate, but the effect fades by week 4. What could explain this and how would you design the test duration?

   MediumStatistics & Experimentation RoundA/B TestingMeta-specific

   Context: Discuss novelty, learning effects, seasonality, and durable impact.

   How to answer: The fading effect suggests novelty effect or user habituation. Initial excitement or curiosity about the new feature drives engagement, but as users become accustomed, their behavior reverts to baseline or they find less utility over time. It could also be due to selection bias if early adopters are more engaged, or a seasonal/external factor. To design the test duration, consider the typical user adoption curve and the natural cycle of the metric, aiming for at least 4-6 weeks to capture habituation and long-term impact, potentially longer if the feature has a cyclical usage pattern.

7. Q7

   In a marketplace-like Facebook Feed feature, treatment users may affect control users. How would network effects or interference bias the experiment?

   MediumStatistics & Experimentation RoundA/B TestingMeta-specific

   Context: Examples include advertiser supply, content inventory, delivery capacity, or pricing pressure.

   How to answer: Network effects, or 'interference,' in an A/B test occur when the treatment group's actions influence the control group's outcomes, or vice-versa. This typically biases the experiment by diluting the true treatment effect, making it appear smaller than it is, or in some cases, amplifying it. For example, if a new feature in treatment causes users to post less, and their friends in control see less content, the control group's engagement might also drop, masking the negative impact of the feature. This violates the Stable Unit Treatment Value Assumption (SUTVA), which is fundamental to valid A/B testing.

8. Q8

   meaningful social interaction rate is a low-frequency event for Instagram Reels. How would you set up an experiment with enough power without waiting too long?

   MediumStatistics & Experimentation RoundA/B TestingMeta-specific

   Context: Discuss proxy metrics, variance reduction, larger samples, longer windows, and risk of metric gaming.

   How to answer: To address low-frequency events like meaningful social interactions (MSI) on Instagram Reels, a strong candidate would propose using a more sensitive, higher-frequency proxy metric that correlates well with MSI. They would then design the experiment around optimizing this proxy, ensuring sufficient sample size and statistical power. Techniques like CUPED or A/B/n testing could further accelerate the experiment by reducing variance or testing more variations simultaneously. Finally, they would emphasize the importance of a follow-up long-term experiment to validate the proxy's impact on the true MSI.

9. Q9

   Design a geo or country-level experiment for WhatsApp Business. When is this better than user-level randomization, and what are the analytical downsides?

   MediumStatistics & Experimentation RoundA/B TestingMeta-specific

   Context: Use matched markets, pre-period balancing, spillover checks, and fewer experimental units.

   How to answer: A geo-level experiment for WhatsApp Business would involve randomizing entire countries or regions to either a treatment or control group for a new feature (e.g., a new business messaging template, an enhanced catalog feature). This is preferable when there are network effects between users within a geography, or when the feature itself is inherently geo-dependent (e.g., regulatory changes, carrier-specific features). The main analytical downsides include lower statistical power due to fewer experimental units, increased sensitivity to pre-existing differences between geos, and potential for spillover effects if businesses or users operate across randomized geographies.

10. Q10

    The Marketplace experiment lifts meaningful social interaction rate overall, but only for new users and only in one country. How would you evaluate heterogeneous treatment effects?

    HardStatistics & Experimentation RoundA/B TestingMeta-specific

    Context: Balance pre-planned segments with exploratory slicing and multiple testing risk.

    How to answer: A strong candidate would first acknowledge the initial finding and the need for subgroup analysis to understand the heterogeneous treatment effects. They would propose methods like CUPED or ANCOVA for variance reduction, and then focus on statistical significance testing within identified subgroups (new users, specific country). They would discuss interaction terms in regression models to formally test for heterogeneity and consider power implications for smaller subgroups. Finally, they would suggest visualizing effects across different dimensions to identify further patterns.

11. Q11

    Treatment improves meaningful social interaction rate but worsens feed load time for Stories. Walk through a launch recommendation

    HardStatistics & Experimentation RoundA/B TestingMeta-specific

    Context: Make a decision under conflicting metrics and quantify tradeoffs for stakeholders.

    How to answer: A strong recommendation balances the trade-offs by quantifying the impact of both metrics. First, assess the statistical significance and practical magnitude of improvement in social interaction vs. degradation in load time. Second, consider the business context and product goals: is Meta prioritizing engagement or performance for Stories right now? Third, propose a phased rollout, A/B/n test with different treatment intensities, or a targeted launch to a segment less sensitive to load time, while continuously monitoring key guardrail metrics. Finally, outline a decision framework for future iterations, potentially exploring engineering solutions to mitigate load time degradation.

12. Q12

    How would you design ramp-up, holdback, and post-launch monitoring for a successful Ads Manager A/B test?

    HardStatistics & Experimentation RoundA/B TestingMeta-specific

    Context: Include ramp stages, persistent holdback, alert thresholds, rollback criteria, and owner accountability.

    How to answer: A strong answer outlines a phased ramp-up strategy, starting with a small percentage of traffic (e.g., 1-5%) to validate technical stability and guardrail metrics, gradually increasing exposure (e.g., 10%, 25%, 50%) based on positive early signals. It then explains the importance of a holdback group (e.g., 1-2% of original traffic) to detect long-term novelty effects or sustained impact post-experiment. Finally, it details post-launch monitoring, focusing on key business metrics, guardrail metrics, and anomaly detection, with clear rollback criteria and ownership.

13. Q13

    Midway through the Facebook Feed test, tracking for WhatsApp Business changed. How would you decide whether the experiment results are still usable?

    HardStatistics & Experimentation RoundA/B TestingMeta-specific

    Context: Compare instrumentation versions, affected traffic share, raw logs, and sensitivity analyses.

    How to answer: A strong candidate would first identify the potential impact of the tracking change on the experiment's validity, specifically concerning external validity and potential confounding. They would propose analyzing the timing of the change relative to the experiment's start and the observed impact on WhatsApp Business metrics in both control and treatment groups. Key steps include checking for a sudden shift or discontinuity in WhatsApp Business metrics post-change, comparing pre- and post-change trends between groups, and assessing if the change disproportionately affected one group or the overall user behavior. If the impact is significant and differential, the results are likely compromised; otherwise, they might still be usable with caveats or by excluding data collected after the change.

14. Q14

    Two overlapping experiments on Instagram Reels both affect ad CTR and revenue per user. How would you detect and manage interaction effects?

    HardStatistics & Experimentation RoundA/B TestingMeta-specific

    Context: Discuss experiment registry, factorial design, exclusion rules, and interaction terms.

    How to answer: A strong candidate would first emphasize the importance of pre-experiment design to minimize overlap, perhaps through orthogonalization or sequential rollout. For detection, they'd suggest analyzing interaction terms in a regression model (e.g., CTR ~ ExpA + ExpB + ExpA*ExpB) or segmenting users by exposure to each experiment and comparing outcomes. Management strategies include re-randomizing to create non-overlapping groups, adjusting power calculations, or using advanced causal inference techniques like CUPED with interaction terms to reduce variance and isolate effects, ultimately informing a decision to pause, adjust, or launch one or both features.

15. Q15

    Meta's Facebook Feed revenue suddenly drops 10% week over week. Structure a business case to diagnose the issue and identify the most likely drivers

    MediumProduct Analytics & Business CaseBusiness CasesMeta-specific

    Context: Consider traffic, conversion, pricing, mix, supply/inventory, outages, marketing, and seasonality.

    How to answer: A strong candidate would structure their diagnosis by first clarifying the 10% drop (e.g., absolute vs. relative, specific product line). They would then propose a framework like 'Internal vs. External' or 'Product vs. User vs. Advertiser' to systematically investigate potential causes. Key areas to explore include recent product changes (algorithm, features), user behavior shifts (engagement, DAU/MAU), advertiser spend changes (budget, bid prices), and external factors (macroeconomic, competitor actions). Finally, they would prioritize investigation based on data availability and potential impact, suggesting specific metrics to monitor for each hypothesis.

Practice these with instant AI feedback in a live mock interview → Start a Meta Data Analyst mock

Be fast and correct on coding; for design, drive the conversation; prepare impact-focused behavioral stories

Indicative Data Analyst pay in India: ~₹6–22 LPA (role-level range, not a Meta-specific figure).

Other roles at Meta

Data Analyst interviews at other companies