Uber Data Analyst Interview Questions (2026)

1. 1Recruiter Screen30 minEasy

   Role targeting, level calibration and process expectations.

2. 2Technical Phone Screen60 minMedium

   One or two medium DSA problems (CodeSignal or live) with emphasis on correct, runnable code and edge cases.

3. 3Onsite Coding I60 minHard

   Practical problem such as building a rate limiter or an in-memory index, judged on working code and API cleanliness.

4. 4Onsite Coding II60 minHard

   Algorithmic problem often with a geospatial or streaming flavor, pushed to optimal complexity.

5. 5System Design60 minHard

   Design a real-time marketplace system (dispatch, ETA, surge) with hard follow-ups on scale, geo-sharding and failure modes.

6. 6Behavioral / Hiring Manager Round45 minMedium

   STAR stories mapped to Uber's cultural norms: ownership, bold bets, customer obsession and conflict handling.

1. Q1

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

   MediumStatistics & Experimentation RoundA/B TestingUber-specific

   Context: Context: Uber wants to match rider demand and driver supply while controlling wait times and cancellations.

   How to answer: A strong answer will define a clear hypothesis, such as 'The new ranking algorithm will increase successful ride requests without negatively impacting driver earnings.' The primary metric should directly reflect this, like 'Successful Ride Request Rate' or 'Conversion Rate from Search to Trip Start.' Guardrail metrics are crucial, including 'Driver Acceptance Rate,' 'Estimated Time of Arrival (ETA),' and 'Cancellation Rate,' to ensure no adverse effects. The randomization unit should be 'User/Rider ID' to maintain consistent experience, and the launch decision rule should involve statistical significance on the primary metric, positive or neutral movement on key guardrails, and a pre-defined minimum detectable effect.

2. Q2

   For Uber Reserve, should randomization happen at rider, session, device, driver, or city level? Explain the tradeoffs

   MediumStatistics & Experimentation RoundA/B TestingUber-specific

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

   How to answer: Randomization for Uber Reserve should primarily happen at the rider level to ensure independent observations and minimize bias, as the feature directly impacts individual rider behavior and booking patterns. Session or device level randomization could lead to contamination if a single rider uses multiple devices/sessions, impacting the same underlying user. Driver level randomization is less suitable as Reserve is a rider-initiated feature, and a driver might serve both experimental and control riders. City level randomization might be considered for market-wide network effects but significantly reduces statistical power and increases cost/time.

3. Q3

   Choose primary and guardrail metrics for a Uber Eats experiment aimed at improving request-to-complete rate. What metrics would prevent a harmful launch?

   MediumStatistics & Experimentation RoundA/B TestingUber-specific

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

   How to answer: A strong candidate would identify 'request-to-complete rate' (or 'conversion rate of requests') as the primary metric, as the experiment directly aims to improve it. For guardrail metrics, they would consider 'average delivery time' and 'customer cancellation rate' to ensure the improvement in completion rate isn't at the expense of user experience. Additionally, 'driver earnings per trip' or 'driver acceptance rate' would be crucial guardrails to prevent negative impacts on the supply side. A harmful launch would be one that improves the primary metric but significantly worsens any of these guardrails, indicating a negative trade-off.

4. Q4

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

   MediumStatistics & Experimentation RoundA/B TestingUber-specific

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

   How to answer: A strong candidate would first define SRM and its implications. They would then propose checking the SRM using a chi-squared test on a pre-experiment population (e.g., all drivers in the eligible pool) to see if the observed split deviates significantly from the expected 50/50. If SRM is confirmed, they would investigate potential causes such as incorrect randomization logic, implementation bugs (e.g., filtering after assignment, client-side assignment issues), or data pipeline errors. Finally, they would discuss the necessary steps to correct the issue and re-run the experiment.

5. Q5

   The Airport Trips 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 TestingUber-specific

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

   How to answer: A strong candidate would explain that peeking early in an A/B test without proper statistical correction inflates the Type I error rate, leading to false positives. They should discuss the need for pre-defined sample sizes and test durations based on power analysis to ensure statistical validity. If early stopping is considered, methods like Sequential Testing (e.g., using an O'Brien-Fleming boundary or a Bonferroni correction for multiple looks) or Bayesian approaches can be mentioned as ways to control error rates. The candidate should emphasize the importance of resisting the urge to stop early due to positive trends without statistical justification, as this can lead to launching features that don't actually provide a lift.

6. Q6

   A new Uber for Business feature shows a large week-1 lift in request-to-complete rate, but the effect fades by week 4. What could explain this and how would you design the test duration?

   MediumStatistics & Experimentation RoundA/B TestingUber-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 a new feature before returning to baseline behavior. Another explanation is selection bias, if the feature was initially rolled out to early adopters who are more engaged, or if the initial lift was driven by a specific segment that completed their 'use case' quickly. External factors like seasonality or concurrent marketing campaigns could also confound results. To design test duration, consider the typical user lifecycle and business cycle for Uber for Business, aiming for at least 2-3 full cycles to capture long-term behavior and account for potential novelty effects or seasonality.

7. Q7

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

   MediumStatistics & Experimentation RoundA/B TestingUber-specific

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

   How to answer: Network effects, or interference, in a marketplace experiment like Uber Rides, typically bias the experiment by diluting the true treatment effect. If treatment users (e.g., those with a new feature that increases demand) compete for the same limited supply as control users, the control group's experience will be negatively impacted, making the treatment appear more effective than it truly is. Conversely, if the treatment reduces demand, control users might benefit from increased supply, making the treatment seem less effective. This leads to an underestimation or overestimation of the feature's actual impact on its target users and the overall marketplace.

8. Q8

   request-to-complete rate is a low-frequency event for Uber Reserve. How would you set up an experiment with enough power without waiting too long?

   MediumStatistics & Experimentation RoundA/B TestingUber-specific

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

   How to answer: To address the low-frequency nature of 'request-to-complete' for Uber Reserve, one must first identify suitable proxy metrics that are higher frequency and highly correlated with the ultimate goal. These proxies could include 'request rate,' 'driver acceptance rate,' or 'pre-trip cancellation rate.' The experiment should then be designed to run longer than typical A/B tests to accumulate sufficient data for the low-frequency primary metric, or a sequential testing approach could be considered. Additionally, consider increasing the sample size by broadening the experiment's geographic scope or user segment if feasible and appropriate, while ensuring the treatment effect is large enough to be detectable given the baseline rate.

9. Q9

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

   MediumStatistics & Experimentation RoundA/B TestingUber-specific

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

   How to answer: A strong candidate would propose a geo-level experiment for Uber Eats by selecting a set of comparable cities or regions, randomizing them into treatment and control groups, and then rolling out the feature within the treatment geos. This approach is superior to user-level randomization when there's a risk of network effects (e.g., changes impacting both eaters and restaurants in a market) or spillover effects that would contaminate user-level results. However, the analytical downsides include reduced statistical power due to fewer experimental units, increased risk of confounding variables if geo-selection isn't robust, and longer experiment durations to detect significant effects.

10. Q10

    The Driver App experiment lifts request-to-complete rate overall, but only for new users and only in one rider_cohort. How would you evaluate heterogeneous treatment effects?

    HardStatistics & Experimentation RoundA/B TestingUber-specific

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

    How to answer: A strong candidate would first identify the need for subgroup analysis, specifically looking at the interaction between treatment, user tenure (new vs. existing), and rider_cohort. They would propose using a regression model (e.g., OLS or logistic regression depending on the outcome variable) with interaction terms to quantify these heterogeneous effects and assess their statistical significance. Key interaction terms would be `treatment * new_user` and `treatment * rider_cohort`. Finally, they would discuss validating these findings and considering potential underlying mechanisms for the observed heterogeneity.

11. Q11

    Treatment improves request-to-complete rate but worsens ETA and cancellation rate for Airport Trips. Walk through a launch recommendation

    HardStatistics & Experimentation RoundA/B TestingUber-specific

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

    How to answer: A strong recommendation would involve segmenting the analysis to understand the trade-offs specifically for Airport Trips and identifying the root causes for the worsened ETA and cancellation rate. Propose a phased rollout or a targeted experiment to mitigate negative impacts, such as geofencing the treatment to specific airport zones or adjusting driver incentives for airport trips. Quantify the net impact by assigning monetary values to each metric (request-to-complete, ETA, cancellations) to determine if the overall business value is positive, considering potential long-term user behavior changes. Finally, recommend a monitoring plan with clear guardrail metrics for ETA and cancellations if the treatment is launched.

12. Q12

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

    HardStatistics & Experimentation RoundA/B TestingUber-specific

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

    How to answer: A strong answer would detail a phased ramp-up strategy, starting with a small percentage of eligible users, closely monitoring key operational metrics and guardrail metrics for anomalies before gradually expanding. For holdback, the candidate should propose a small, stable percentage of the target population (e.g., 1-5%) to remain on the control experience indefinitely, serving as a long-term baseline for assessing sustained impact and novelty effects. Post-launch monitoring involves continuous tracking of primary metrics, secondary metrics, and operational health, establishing clear alert thresholds and an incident response plan, and scheduling regular reviews to detect regressions or unexpected long-term shifts.

13. Q13

    Midway through the Rides test, tracking for Uber Eats changed. How would you decide whether the experiment results are still usable?

    HardStatistics & Experimentation RoundA/B TestingUber-specific

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

    How to answer: First, analyze the nature and timing of the tracking change for Uber Eats relative to the Rides test. Determine if the change introduced a systemic bias or increased variance in the key metrics for the Rides experiment. Compare pre- and post-change trends and distributions for both control and treatment groups, focusing on Rides metrics and potential spillover effects. If the impact is negligible, quantifiable, and correctable, the results might be usable; otherwise, the experiment should be invalidated or restarted.

14. Q14

    Two overlapping experiments on Uber Reserve both affect net revenue per trip. How would you detect and manage interaction effects?

    HardStatistics & Experimentation RoundA/B TestingUber-specific

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

    How to answer: A strong candidate would first identify the need for a statistical test to detect interaction effects, such as a multi-factor ANOVA or regression model including an interaction term. They would then propose segmenting users into four groups (Control, Exp A only, Exp B only, Exp A+B) and analyzing the net revenue per trip for each group, comparing the observed combined effect to the sum of individual effects. To manage interactions, they would suggest either sequential deployment, mutually exclusive user populations, or a joint decision-making framework based on the magnitude and direction of the interaction. Finally, they would emphasize the importance of pre-experiment power analysis to detect expected interaction effects.

15. Q15

    Uber's Rides 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 CasesUber-specific

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

    How to answer: A strong candidate would first clarify the scope (e.g., global vs. specific region/product) and then structure their diagnosis into three main categories: internal factors (product/pricing changes, operational issues, tech bugs), external factors (competitor actions, regulatory changes, seasonality/events, economic shifts), and data issues (tracking errors, reporting glitches). They would propose a systematic drill-down, starting with high-level metrics (trips, average fare, active riders, retention) and segmenting by geography, rider type, and product tier. Finally, they would prioritize investigation based on potential impact and ease of data access, recommending specific data points to analyze for each hypothesis.

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

Strong DSA and scalable system design; prepare analytical/behavioral stories

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

Other roles at Uber

Data Analyst interviews at other companies