Walmart Labs (Walmart Global Tech) Data Analyst Case Interview — Omnichannel Metrics, Experimentation, and SQL

This case mirrors real Walmart Labs data analyst interviews where you work through a practical e-commerce and stores-to-online problem at Walmart scale. You are embedded with an omnichannel team (e.g., Online Pickup & Delivery or Marketplace) and asked to: 1) define north-star and guardrail metrics tied to Walmart’s customer-first and Everyday Low Price ethos (conversion, add-to-cart rate, order fill rate, on-shelf availability proxy, cancellations, on-time delivery, substitution impact, margin/AOV trade-offs); 2) write performant SQL against large behavioral and transactional datasets (clickstream events, orders, line items, inventory snapshots, store attributes, experiment assignments) using CTEs, window functions, and careful joins to avoid double counting; 3) analyze an A/B test or pre/post rollout (e.g., a pickup-substitution preference feature) including experiment design, power/significance, novelty effects, bucketing integrity, and guardrails (latency, defect rate, NPS); 4) run a root-cause deep dive on a metric drop by slicing cohorts (device, store type, region, new vs. repeat, delivery vs. pickup), checking data quality/instrumentation, and quantifying business impact; 5) propose a monitoring/dashboard plan (KPIs, alerts, SLAs) and next steps with clear stakeholder communication (PMs, operations, engineers). Interviewers emphasize structured thinking, bias for action, scale-aware SQL, and crisp trade-off narration aligned with Walmart’s values of service to the customer and operational excellence.