Walmart Data Analyst Interview Questions (2026)

Online coding test -> 2-3 technical rounds (DSA, system design) -> hiring manager

1. Q1

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

   MediumStatistics & Experimentation RoundA/B TestingWalmart-specific

   Context: Context: Walmart wants to grow omnichannel adoption while keeping fulfillment reliable and inventory visible.

   How to answer: A strong answer will define a clear, testable hypothesis, such as 'The new ranking algorithm will increase add-to-cart rate by 2% without negatively impacting session length.' The primary metric should directly reflect the hypothesis, like 'add-to-cart rate per user.' Guardrail metrics are crucial to detect negative side effects, such as 'average session duration' or 'app crash rate.' The randomization unit should be carefully chosen, likely 'user ID' to maintain a consistent experience. Finally, a clear launch decision rule, such as 'statistically significant (p<0.05) positive lift in primary metric and no statistically significant negative impact on guardrails over two full weeks,' should be established.

2. Q2

   For Walmart+, should randomization happen at customer, session, device, marketplace seller, or market level? Explain the tradeoffs

   MediumStatistics & Experimentation RoundA/B TestingWalmart-specific

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

   How to answer: Randomization for Walmart+ should primarily happen at the customer level. This ensures that a single customer consistently experiences either the control or treatment, preventing contamination and ensuring independent observations. While session or device level might seem easier to implement, they risk a single customer seeing both versions, leading to skewed results. Marketplace seller or market level randomization is generally too broad for a customer-centric product like Walmart+, unless the experiment specifically targets seller behavior or market-specific features.

3. Q3

   Choose primary and guardrail metrics for a Online Grocery experiment aimed at improving digital-to-store conversion. What metrics would prevent a harmful launch?

   MediumStatistics & Experimentation RoundA/B TestingWalmart-specific

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

   How to answer: For a digital-to-store conversion experiment, the primary metric should directly measure the conversion from online interaction to in-store action, such as 'Online Grocery Order Placed (Digital) to In-Store Pickup/Delivery Completion Rate'. Guardrail metrics are crucial to prevent negative side effects. Key guardrails would include 'Average Order Value (AOV)', 'Number of Orders Placed', 'Customer Satisfaction Score (CSAT) for Online Grocery', and 'App/Website Engagement (e.g., sessions, time spent)'. These guardrails ensure the experiment doesn't cannibalize other revenue streams, reduce overall order volume, degrade customer experience, or negatively impact platform usage.

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 TestingWalmart-specific

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

   How to answer: A strong candidate would first define Sample Ratio Mismatch (SRM) and state its importance. Then, they would outline a diagnostic approach: calculate the expected vs. observed split, perform a chi-square goodness-of-fit test to determine statistical significance, and check for randomization issues. Finally, they would discuss potential root causes like implementation bugs, traffic routing problems, or user ID inconsistencies.

5. Q5

   The Store Pickup 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 TestingWalmart-specific

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

   How to answer: Explain that peeking early in an A/B test without pre-planned sequential testing methods inflates Type I error, leading to false positives. Describe how sequential testing (e.g., using O'Brien-Fleming boundaries or Always Valid p-values) allows for valid early stopping by adjusting significance thresholds at pre-determined or continuous check points. Advise against stopping early based on a single peek, recommending to either let the experiment run to its pre-calculated duration or, if early stopping is critical, to re-evaluate with a sequential testing framework in mind for future experiments. Emphasize the importance of statistical rigor and avoiding premature conclusions to ensure reliable business decisions.

6. Q6

   A new Retail Media feature shows a large week-1 lift in digital-to-store conversion, but the effect fades by week 4. What could explain this and how would you design the test duration?

   MediumStatistics & Experimentation RoundA/B TestingWalmart-specific

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

   How to answer: The fading effect could be due to novelty effect, where initial user excitement or curiosity drives engagement, but sustained value is lacking. Another explanation is a 'honeymoon period' with early adopters, or a limited pool of users for whom the feature is highly relevant, leading to saturation. External factors like seasonality or concurrent promotions could also confound results. To design test duration, analyze historical data for similar feature adoption curves and seasonality, and consider the feature's expected long-term impact and user lifecycle. A minimum of 4-6 weeks is often a good starting point to observe stabilization and account for weekly cycles, potentially extending to 8-12 weeks if long-term behavioral shifts are expected.

7. Q7

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

   MediumStatistics & Experimentation RoundA/B TestingWalmart-specific

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

   How to answer: Network effects, or 'interference,' bias A/B tests by violating the Stable Unit Treatment Value Assumption (SUTVA), meaning a user's outcome is not solely dependent on their own treatment status. In a marketplace, treatment users (e.g., those seeing a new feature) might influence control users' behavior by changing inventory availability, pricing, or visibility of items. This leads to an underestimation or overestimation of the true treatment effect, as the control group's baseline is no longer representative of what it would be without the treatment group's presence. The observed difference between groups would then reflect both the direct treatment effect and the indirect interference effect, making it difficult to isolate the true impact of the feature.

8. Q8

   digital-to-store conversion is a low-frequency event for Walmart+. How would you set up an experiment with enough power without waiting too long?

   MediumStatistics & Experimentation RoundA/B TestingWalmart-specific

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

   How to answer: To address low-frequency conversion, I would first define a suitable proxy metric that is highly correlated with digital-to-store conversion but occurs more frequently, such as 'adding an in-store item to cart' or 'viewing store-specific inventory'. I would then consider using a sequential testing approach, like an A/B test with early stopping rules, to potentially conclude the experiment faster if a significant difference emerges. Another strategy is to increase the sample size significantly by expanding the experiment's reach (e.g., more geo-locations, longer duration if feasible within time constraints) or by segmenting users more broadly if the treatment effect is expected to be uniform. Finally, I would ensure the minimum detectable effect (MDE) is set realistically, acknowledging that a smaller MDE requires more data and thus more time.

9. Q9

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

   MediumStatistics & Experimentation RoundA/B TestingWalmart-specific

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

   How to answer: A geo-level experiment for Online Grocery would involve randomizing entire geographic markets (e.g., DMAs, counties, or zip codes) to either a treatment or control group. This is superior to user-level randomization when there are network effects, spillover effects, or when the intervention itself is difficult or impossible to implement at an individual user level (e.g., changes to pricing algorithms, delivery zones, or marketing campaigns that impact an entire region). Analytical downsides include lower statistical power due to fewer experimental units, increased risk of selection bias if geo units are not truly comparable, and challenges in controlling for confounding variables that vary by geography.

10. Q10

    The Marketplace experiment lifts digital-to-store conversion overall, but only for new users and only in one category. How would you evaluate heterogeneous treatment effects?

    HardStatistics & Experimentation RoundA/B TestingWalmart-specific

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

    How to answer: To evaluate heterogeneous treatment effects (HTE), I would first define relevant subgroups based on user tenure (new vs. existing) and product categories. Then, I would conduct separate A/B tests or analyze the existing experiment data within each subgroup, focusing on interaction terms in regression models to quantify the differential impact. Techniques like Causal Forests or Bayesian Additive Regression Trees (BART) could be employed for more complex HTE discovery, followed by validation of findings through further experimentation or observational studies where feasible. Finally, I would interpret the practical significance of these subgroup-specific lifts for business strategy.

11. Q11

    Treatment improves digital-to-store conversion but worsens substitution and pickup delay rate for Store Pickup. Walk through a launch recommendation

    HardStatistics & Experimentation RoundA/B TestingWalmart-specific

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

    How to answer: A strong recommendation requires quantifying the trade-offs. Calculate the monetary value of the conversion improvement against the cost of increased substitution and pickup delays. Consider segmenting users to apply the treatment only where net positive, or to mitigate negative impacts. Propose a phased rollout with clear monitoring metrics and a rollback plan, focusing on long-term customer value and operational efficiency.

12. Q12

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

    HardStatistics & Experimentation RoundA/B TestingWalmart-specific

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

    How to answer: A strong candidate would design ramp-up by gradually increasing traffic to the experiment, starting with a small percentage (e.g., 1-5%) and monitoring key metrics for stability and sanity checks before expanding. Holdback would involve reserving a small, representative portion of the eligible user base (e.g., 0.5-1%) that never sees the new feature, serving as a long-term control for potential novelty effects or systemic biases. Post-launch monitoring requires establishing a dashboard with pre-defined success metrics (e.g., ROAS, CTR, conversion rate) and guardrail metrics (e.g., page load time, error rates) with automated alerts, continuously comparing treatment to control for sustained impact and unexpected regressions.

13. Q13

    Midway through the Walmart App test, tracking for Online Grocery changed. How would you decide whether the experiment results are still usable?

    HardStatistics & Experimentation RoundA/B TestingWalmart-specific

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

    How to answer: First, identify the exact nature and timing of the tracking change for Online Grocery relative to the experiment's start and the change point. Then, assess if the change affects both the control and treatment groups equally and if it impacts the primary success metrics. If the change occurred before the experiment started or affects both groups symmetrically and doesn't invalidate the metric definition, the results might still be usable with careful interpretation. Otherwise, consider segmenting the data to analyze pre-change and post-change periods separately, or, if the impact is severe and asymmetrical, the experiment may need to be restarted.

14. Q14

    Two overlapping experiments on Walmart+ both affect gross margin dollars. How would you detect and manage interaction effects?

    HardStatistics & Experimentation RoundA/B TestingWalmart-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, ideally by segmenting user populations or using different launch times. If overlap is unavoidable, they would propose using a statistical model (e.g., regression) that includes an interaction term between the two experiment treatment indicators to quantify the interaction effect. They would then discuss methods for managing the interaction, such as prioritizing one experiment, re-designing the experiments to be mutually exclusive, or accepting a more complex interpretation of results. Finally, they would highlight the need for robust monitoring of key metrics during the experiment to detect unexpected shifts.

15. Q15

    Walmart's Walmart App 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 CasesWalmart-specific

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

    How to answer: A strong business case would start by clarifying the problem (magnitude, timing, scope) and forming initial hypotheses across internal and external factors. The candidate should then outline a structured diagnostic approach, starting with data validation and then segmenting the revenue drop by key dimensions like user segment, product category, geography, device, and time of day. Finally, they should propose potential root causes and data points to investigate for each, such as app performance issues, marketing campaign changes, competitor actions, or inventory problems.

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

Practise DSA and system design for scale; prepare real project deep-dives; expect large-scale scenario questions

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

Other roles at Walmart

Data Analyst interviews at other companies