Top Zomato Product Manager Interview Questions 2026

Diagnose root causes first: restaurant-side (delay, item out of stock, wrong order), delivery-side (partner cancel, long wait), customer-side (changed mind, slow ETA). Interventions per cause: better restaurant ETA prediction, inventory sync APIs, partner incentives to honor accepts, customer-facing transparency on prep status. Measure cancellation rate segmented by reason — single number hides the levers.

Score candidates on proximity + availability + rating + current load + direction match (heading toward restaurant). Real-time dispatch with 30-second timeout; on rejection, escalate score weights for incentives. Geohash-based candidate retrieval. Surge incentives during high-demand periods. Fairness layer prevents one partner from getting all premium orders. Batch optional matching every N seconds to optimize globally.

STAR with specific data gaps (no historical baseline, small sample, biased segment). Action: enumerated assumptions, picked a reversible decision, built in a checkpoint. Outcome: shipped on time, validated post-launch, iterated when assumptions broke. Show comfort with calculated risk rather than analysis paralysis.

Multi-layered. North star: weekly active buyers (or order frequency per active user). Operational: delivery time (10-min promise compliance), item fill rate (in-stock at order). Quality: NPS, return rate. Financial: contribution margin per dark store, GMV per dark store. Counter-metrics: cancellation rate, customer complaint volume. Cohort retention to validate stickiness.

Supply-side seeding first: onboard 200-300 restaurants spanning popular cuisines + price points. Demand activation: heavy discounts for first 4 weeks, local marketing partnerships, college outreach. Delivery partner recruitment: targeted hiring with bike-purchase finance. Local payment methods (UPI dominant). Hyperlocal ops team for issue resolution. Measure: 90-day retention, restaurant churn, partner earnings, GMV growth curve.

Pick a feature you actually use. Walk through one user pain you've personally felt. Propose a concrete data-driven improvement tied to a metric. Avoid vague "make it faster/prettier" answers — be specific about the user segment, the behavior change, and the success metric you'd watch.

Step 1: rule out a data/instrumentation bug. Step 2: segment by city/platform/cohort/new-vs-existing. Step 3: external — competitor launch (Swiggy promo?), policy change, weather, festivals. Step 4: funnel analysis — is acquisition down or retention/engagement? Step 5: form a hypothesis, validate with the smallest possible signal, propose mitigation.

Tiered benefits aligned to behavior: free delivery, priority support, exclusive restaurant deals, early access to drops. Subscription pricing tested against perceived value. Retention focus: churn prediction triggers (lapsed users get bonus credits). Unit economics: ensure incremental orders cover the discount cost. Cohort cohorts to measure LTV uplift. Avoid pure discount race — gamify status and exclusivity.

Quantify the impact on retention, frequency, and NPS for each. Identify the current bottleneck via funnel analysis — is drop-off at search or post-order satisfaction? Frame as A/B testable bets. LTV analysis: a 10% speed improvement may unlock more order frequency than discovery tweaks. Decide with data, but also consider strategic moat (delivery speed is harder for competitors to copy in saturated markets).

Cross-functional story showing operational depth. Frame the product issue (e.g., long ETAs in dense areas), how you engaged ops to understand ground reality (rode with delivery partners, visited dark stores), and how the combined product+ops solution worked. Numbers — ETA reduced by X minutes, complaint volume down Y%.

Predict daily surplus by combining historical orders, weather, day-of-week, and local events. Restaurant dashboard alerts on over-production risk. Discounted "last-2-hour" listing surfaced in app to nearby users. Opt-in for restaurants; freshness guarantees clearly labeled. Measure: waste reduced (kg), incremental revenue captured, NPS impact, repeat order rate from waste-discount buyers.

Issues with current systems: ratings biased by sample (only frustrated/delighted users rate), no recency weighting, fake reviews, no delivery-vs-dine-in separation. Improvements: time-decayed ratings, separate dimensions (food, packaging, delivery time), fake-review detection (account age, review velocity), restaurant response feature for owner accountability. Show how each change affects fairness and signal quality.

GMV = total order value flowing through the platform. Revenue = what Zomato actually earns: commissions from restaurants (15-25% take rate), delivery fees from customers, ads paid by restaurants, Gold subscription, Hyperpure margin. Take rate = Revenue / GMV. Unit economics matter — high GMV with thin margin is fragile.

Cold chain logistics — refrigerated trucks, temperature monitoring. Supplier onboarding with quality audits and dairy-specific certifications. Demand forecasting per restaurant by historical orders + weather. Procurement portal with bulk ordering, delivery schedules, returns for spoilage. Freshness guarantees (e.g., delivered within 24h of milking). Pricing model: subscription or per-order with volume discounts.

Show prioritization maturity. Frame the feature, why you built it, the metrics that flagged failure (no adoption, negative side effects, opportunity cost). Walk through the sunset decision — data-driven, communicated to stakeholders, learnings documented. End with what got built instead and the resulting impact. Killing features cleanly is a senior PM trait.