Why Did Zomato's Restaurant Ratings Drop?

Zomato's average restaurant rating in Pune fell 10% month-over-month. The decline was sharp, sudden, and confined to a single city — exactly the kind of signal that demands a structured investigation before a fix.

Through a disciplined Q&A process, the root cause was isolated to a single policy change — the retraction of a delivery discount code previously available to non-Gold subscribers — compounded by the arrival of a new local tiffin-delivery competitor. The takeaway: a pricing decision, not a product bug or service failure, drove perceived value down and review behaviour with it.

Before generating hypotheses, the metric itself was stress-tested. Three constraints fell out of the clarifying questions and narrowed the search space dramatically.

• → Calculation unchanged — straight average of visible restaurant ratings, no methodology shift.
• → Geography-specific — drop confined to Pune; other cities unaffected.
• → Sudden, not gradual — appeared within a single month vs. the prior month.
• → Segment-specific — driven by non-Gold members, not the Gold cohort.
• → Channel-specific — exclusive to food delivery; dine-in unaffected.
• → Category-agnostic — fine dining, fast food, and local cuisine all dropped equally.

A structured hypothesis tree split possible causes into three buckets — External Events, Internal Events, and Operational Changes — so no category was skipped before committing to a path.

External factors: a new local tiffin-delivery app had entered Pune, giving consumers an affordable everyday alternative. No major events, negative press, regulatory shifts, or seasonal effects.

Internal product changes: only one material signal — a delivery discount code valid across major Pune restaurants had been withdrawn for users without a Zomato Gold membership. App updates, catalogue changes, and the rating flow itself were all clean.

Tech & infrastructure: no bugs, no version-specific issues, no backend migrations. Operational KPIs — order volume, delivery executives, customer service quality, delivery times — were all stable.

The standard post-delivery rating journey was walked end-to-end to confirm no UX friction was suppressing reviews:

• → User receives a push notification prompting them to rate the delivery.
• → Alternative path: Open app → Profile → Order History → Select order → Review Order.
• → Rate the restaurant (1–5 stars).
• → Optionally add a written review and/or photo.
• → Tap Submit.

"After ruling out tech bugs, operational failures, and most external events, two signals remained: the retraction of delivery discount codes for non-Gold users, and the arrival of a competing tiffin-delivery app. Independently lowering perceived value — together, amplifying dissatisfaction."

Cause 1 — Retraction of discount codes for non-Gold users. The rating drop aligned exactly with this segment, making the causal link strong. Three validation passes were proposed:

• → Order & rating pattern analysis — frequency and average stars for non-Gold users, month before vs. month after the retraction.
• → User feedback mining — app reviews, in-app feedback, and support tickets for mentions of 'offers', 'discounts', or 'Gold'.
• → Rating time-series correlation — daily average rating for non-Gold users plotted against the discount removal date, looking for a step-change.

A new tiffin-delivery app in Pune may have reset users' value benchmark. When a cheaper alternative exists, willingness to overlook imperfections falls — and that surfaces as lower ratings.

• → Market share & migration analysis — overlap between Zomato non-Gold users and the competitor's user base.
• → CSAT & NPS comparison — competitor satisfaction data vs. Zomato's Pune scores, by attribute.
• → Competitor promotional audit — current offers, pricing, and discount depth that could be undercutting Zomato's non-Gold experience.

Solution 1 — Reintroduce targeted delivery incentives without reversing the policy wholesale, protecting margin while restoring perceived value:

• → Tiered discount codes — smaller for non-Gold, deeper for Gold, creating a clear upgrade incentive.
• → Loyalty reward credits — cashback or Zomato Credits that encourage repeat ordering instead of direct discounts.
• → Restaurant-funded offers — co-sponsored delivery discounts with high-volume Pune restaurants.
• → Transparent communication — push + email explaining the change and the alternative value users can access.

To blunt the competitor's impact and defend share in Pune:

• → Targeted campaigns highlighting Zomato's catalogue depth, reliability, and restaurant variety — attributes the new entrant can't match yet.
• → Partner with restaurants to elevate delivery standards in Pune — packaging quality and delivery times must stay superior.
• → Hyperlocal marketing in Pune neighbourhoods where the competitor has the strongest footprint.

Shipping a fix is half the job. The monitoring layer makes sure the recovery is real and surfaces the next anomaly before it compounds.

Three operational habits to catch the next shift early:

• → Real-time rating alerts — automated triggers on ≥2% daily moves in any city × segment × channel combination.
• → Monthly post-delivery CSAT surveys for non-Gold Pune users — attitudinal shifts surface before star ratings move.
• → Quarterly competitive benchmarking — pricing, catalogue, delivery times, and promotional activity.

A 10% drop in an aggregate rating sounds systemic. Disciplined investigation revealed a narrow, identifiable cause: a single discount-policy decision, amplified by a competitor's arrival that reset users' value benchmark.

The resolution path is equally narrow — restore perceived value through smart incentive design, defend competitive positioning through quality and targeted marketing, and instrument the monitoring needed to catch the next anomaly inside a week instead of a month.