Google Business Analyst Interview Questions (2026)

1. 1Recruiter screen30 minEasy

   Background, level calibration, and process walkthrough with a recruiter.

2. 2Technical phone screen45 minHard

   One or two DSA problems solved live in a shared editor with emphasis on optimal complexity and clean code.

3. 3Coding round (onsite)45 minHard

   Harder DSA with follow-up constraint changes; interviewer scores GCA and RRK on a rubric.

4. 4System design round45 minHard

   Design a planet-scale system (e.g. a piece of Search or YouTube) with explicit capacity estimates and tradeoffs.

5. 5Googleyness & Leadership45 minMedium

   Behavioral round on collaboration, ambiguity, and user-first judgment scored against Google's structured rubric.

6. 6Hiring Committee review30 minMedium

   No candidate interaction; the written feedback packet is reviewed and the hire/no-hire decision is made, followed by team matching.

1. Q1

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

   MediumStatistics & Experimentation RoundA/B TestingGoogle-specific

   Context: Context: Google wants to improve relevance while protecting user trust and privacy.

   How to answer: A strong answer will define a clear, testable hypothesis for the Search ranking change (e.g., 'New algorithm increases CTR'). The primary metric should directly reflect the hypothesis, such as Click-Through Rate (CTR) or query success rate, with guardrail metrics like latency, error rate, and revenue to prevent negative side effects. The randomization unit should be the user or search session to ensure independent observations, and the launch decision rule should specify statistical significance thresholds for the primary metric and acceptable ranges for guardrails over a defined test duration.

2. Q2

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

   MediumStatistics & Experimentation RoundA/B TestingGoogle-specific

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

   How to answer: Randomization for YouTube Shorts should primarily happen at the user level to ensure independent observations and avoid contamination. Session or device level randomization could lead to a single user experiencing both control and experiment, biasing results. Advertiser or country level randomization is generally too coarse for product features like Shorts, as it would mask individual user behavior and require massive sample sizes for statistical significance, though it might be relevant for specific ad-tech or geo-specific feature tests. The key tradeoff is between minimizing contamination and achieving sufficient statistical power efficiently.

3. Q3

   Choose primary and guardrail metrics for a Google Ads experiment aimed at improving query success rate. What metrics would prevent a harmful launch?

   MediumStatistics & Experimentation RoundA/B TestingGoogle-specific

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

   How to answer: A strong candidate would identify 'Query Success Rate' (e.g., clicks on an ad, conversion after ad click) as the primary metric. For guardrail metrics, they would propose 'Revenue per User/Query' and 'Ad Impressions/Clicks per User' to ensure the experiment doesn't negatively impact monetization or user engagement with ads. Additionally, 'Latency' or 'Error Rate' could be guardrails to prevent system degradation. The metrics preventing a harmful launch are primarily the monetization and core engagement guardrails, ensuring no significant drop in revenue or user interaction with the ads.

4. Q4

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

   MediumStatistics & Experimentation RoundA/B TestingGoogle-specific

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

   How to answer: A strong candidate would first confirm the SRM by checking the observed split against the expected split using a chi-squared test. Next, they would investigate potential causes, starting with the random assignment mechanism itself (e.g., hash function issues, caching problems, or user ID inconsistencies). They would also consider external factors like bot traffic, data logging errors, or pre-existing user segments that might disproportionately fall into one group. Finally, they would discuss the implications of SRM on experiment validity and how to mitigate it, such as re-running the experiment or adjusting for the mismatch if possible.

5. Q5

   The Google Play 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 TestingGoogle-specific

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

   How to answer: A strong candidate would explain that stopping an A/B test early due to positive trending results, also known as 'peeking,' inflates the Type I error rate (false positive rate). They would discuss the importance of pre-determining sample size and test duration based on statistical power and minimum detectable effect (MDE) to maintain statistical validity. The candidate should recommend continuing the experiment until the pre-determined end date or sample size is reached, or employing sequential testing methodologies if early stopping is truly desired, which requires specific statistical adjustments. They would emphasize the risk of launching a feature that might not actually be positive in the long run, leading to wasted resources and potential negative user impact.

6. Q6

   A new Cloud Marketplace feature shows a large week-1 lift in query success rate, but the effect fades by week 4. What could explain this and how would you design the test duration?

   MediumStatistics & Experimentation RoundA/B TestingGoogle-specific

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

   How to answer: The fading effect could be due to novelty effect, where users initially engage more with the new feature but revert to prior behavior once the novelty wears off. Another explanation is a selection bias, where early adopters (who are often more tech-savvy or engaged) are the first to try the feature, inflating initial metrics. The feature might also have short-term benefits that don't translate to long-term value, or there could be external factors or seasonality influencing the later weeks. To design the test duration, I would recommend a minimum of 4-6 weeks to capture potential novelty effects and allow for user habituation. Longer durations (8-12 weeks) would be ideal to observe sustained impact, account for weekly/monthly cycles, and ensure statistical significance over time, especially for features impacting long-term user behavior or revenue.

7. Q7

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

   MediumStatistics & Experimentation RoundA/B TestingGoogle-specific

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

   How to answer: Network effects in a marketplace Search feature mean that the actions of treatment users (e.g., increased engagement or purchases due to a new feature) can directly or indirectly influence control users. This interference primarily biases the experiment by understating the true treatment effect if control users benefit from the treatment group's actions (positive spillover), or overstating it if control users are negatively impacted (negative spillover). For example, if treatment users consume more inventory, control users might see less availability, making the control group perform worse than it would in isolation. Conversely, if treatment users create more supply, control users might benefit from increased options, making the control group perform better.

8. Q8

   query success rate is a low-frequency event for YouTube Shorts. How would you set up an experiment with enough power without waiting too long?

   MediumStatistics & Experimentation RoundA/B TestingGoogle-specific

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

   How to answer: To set up an A/B test for a low-frequency event like query success rate on YouTube Shorts without excessive waiting, the primary strategy involves increasing the sample size significantly or extending the experiment duration if feasible. Alternatively, one could consider using a more sensitive proxy metric that correlates with query success but occurs more frequently, such as 'query initiated' or 'search results page viewed'. Another approach is to employ a sequential testing methodology, allowing for early stopping if a significant effect is observed, or to utilize CUPED (Controlled-experiment Using Pre-experiment Data) to reduce variance and thus the required sample size.

9. Q9

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

   MediumStatistics & Experimentation RoundA/B TestingGoogle-specific

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

   How to answer: A geo-level experiment for Google Ads would involve randomizing entire geographic regions (e.g., states, countries, DMAs) into treatment and control groups, rather than individual users. This approach is superior to user-level randomization when the treatment has network effects, spillover, or requires changes to backend systems that are difficult to roll out partially. For Google Ads, this could apply to changes impacting advertiser bidding behavior, ad quality algorithms, or new ad formats that might influence an entire market. Key analytical downsides include lower statistical power due to fewer experimental units, increased sensitivity to selection bias if geos are not well-matched, and challenges in interpreting results due to confounding factors unique to specific regions.

10. Q10

    The Google Maps experiment lifts query success rate overall, but only for new users and only in one device_type. How would you evaluate heterogeneous treatment effects?

    HardStatistics & Experimentation RoundA/B TestingGoogle-specific

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

    How to answer: A strong candidate would first acknowledge the initial positive overall result and then immediately identify the need to segment the data by user type (new vs. existing) and device type to confirm the observed heterogeneity. They would propose running separate A/B tests or subgroup analyses within the existing experiment for each segment, focusing on the 'new users' and the specific 'device_type' where the lift was observed. The evaluation would involve calculating statistical significance and effect size for these specific subgroups, comparing them to the control group within those same segments, and considering potential reasons for the differential impact, such as UI changes being more intuitive for new users or specific device capabilities. Finally, they would discuss the implications for launch strategy, potentially rolling out the feature only to the successful segment initially.

11. Q11

    Treatment improves query success rate but worsens page load latency for Google Play. Walk through a launch recommendation

    HardStatistics & Experimentation RoundA/B TestingGoogle-specific

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

    How to answer: A strong recommendation would involve a phased rollout strategy, starting with a small, controlled segment to gather more data on the trade-offs. Key considerations include quantifying the user impact of both metrics (e.g., how much more revenue from successful queries vs. how much abandonment due to latency), analyzing user segments (e.g., users with fast vs. slow connections), and exploring technical mitigations for latency. The recommendation should propose a clear decision framework based on weighted metrics and potential follow-up experiments to optimize the user experience.

12. Q12

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

    HardStatistics & Experimentation RoundA/B TestingGoogle-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 detect immediate negative impacts, followed by gradual increases (e.g., 10%, 25%, 50%) while continuously monitoring key metrics. For holdback, the candidate should propose retaining a small, untreated control group (e.g., 1-2%) even after 100% rollout, to measure long-term incrementality and guard against novelty effects or seasonal changes. Post-launch monitoring involves establishing a dashboard with critical business and technical metrics (e.g., conversion rates, revenue, error rates, latency) with clear alert thresholds, ensuring ongoing performance validation and rapid issue detection.

13. Q13

    Midway through the Search test, tracking for Google Ads changed. How would you decide whether the experiment results are still usable?

    HardStatistics & Experimentation RoundA/B TestingGoogle-specific

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

    How to answer: A strong candidate would first identify the specific change in Google Ads tracking and its potential impact on the Search experiment's primary metrics (e.g., clicks, impressions, conversions). They would then propose analyzing pre- vs. post-change data for both control and experiment groups to detect any significant shifts or divergences in trends. Key steps include checking for parallel trends before the change, assessing the magnitude and direction of the impact, and determining if the change disproportionately affected one group or the overall experiment validity. Finally, they would conclude whether the data is still usable, potentially with adjustments or a restart, based on the analysis.

14. Q14

    Two overlapping experiments on YouTube Shorts both affect ad revenue per mille. How would you detect and manage interaction effects?

    HardStatistics & Experimentation RoundA/B TestingGoogle-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-analysis, including understanding the experimental designs and potential overlap points. They would then propose statistical methods like ANCOVA or regression analysis with interaction terms to detect significant interaction effects on ad revenue per mille. Mitigation strategies would involve sequential rollout, re-designing experiments to be mutually exclusive, or using a switchback/time-series approach if direct user segmentation is impossible. Finally, they would discuss the trade-offs of each approach in terms of statistical power, time, and complexity.

15. Q15

    Google's Search 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 CasesGoogle-specific

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

    How to answer: A strong business case would first define the problem (10% WoW Search revenue drop) and its potential impact. The analysis should then segment the problem by geography, device, user type, and ad type to pinpoint where the drop is most significant. Hypotheses should be generated across internal factors (e.g., algorithm changes, ad platform bugs, sales team performance) and external factors (e.g., competitor actions, economic shifts, seasonality). Finally, a recommendation for data-driven investigation and potential mitigation strategies should be proposed, prioritizing based on impact and ease of implementation.

Practice these with instant AI feedback in a live mock interview → Start a Google Business Analyst mock

Master DSA and communicate your thinking out loud; use Google's structured Explain-Clarify-Improve approach; prepare for Googleyness/behavioral

Indicative Business Analyst pay in India: ~₹7–26 LPA (role-level range, not a Google-specific figure).

Other roles at Google

Business Analyst interviews at other companies