Zomato Android redesign holds up

I opened an old Zomato for Android redesign I did in 2016. The pixels aged, sure, but the assumptions are the real story. The interesting bit is which ones still hold up in 2025. 🍽️👇

Found a note dated Mar 25, 2017 in my messy notes folder that just said “Publish Zomato Medium article”. I wrote the piece in 2016 and hit publish a year later. Context then: Material Design 1.0, Android 5 to 7, Jio’s 4G wave taking off, and Zomato mostly about restaurant discovery.

2016 core bet and constraints I designed around:
- Core bet: help me decide where to eat nearby in under a minute
- Navigation: move from drawer to bottom nav, keep search one tap away, map visible but list is the fastest path
- Performance: skeletons over spinners, cache hard for flaky networks, keep time to first result low
- Permissions: location as a clear value exchange, ask with runtime prompts and context

What aged well vs poorly:
- Aged well: bottom navigation, skeleton loading, offline-friendly caching
- Mixed: photo-first cards vs info density for quick decisions
- Aged poorly: heavy custom UI across OEM variants, growth nudges around location and notifications during rising privacy fatigue

2025 reality that reframes the work:
- Platform: Material You and Material 3, dynamic color, predictive back, Compose by default, notifications gated, background location tighter, Compose changes the cost of custom components
- Product: high-frequency delivery and logistics, trust and safety, ETA clarity, and reliability matter more than “delight”
- Payments and flows: UPI rails and one-tap reviews shift conversion math, server-driven UI and feature flags move faster than static IA

If I did it today:
- Start with a system: tokens, theming, motion
- Spend custom UI only where it truly differentiates
- Budget cold-start latency like a feature
- Make ranking and personalization explainable
- Treat permissions as long-term trust contracts

Open questions I’m still chewing on:
- How should discovery UIs expose ranking logic without hurting conversion
- Where do LLMs help most - query rewrite, venue summaries, or intent modeling - and how do we keep user agency
- Can Compose plus server-driven UI slow design drift at scale

Here’s the original note from 2017: https://github.com/AnandChowdhary/notes/blob/main/notes/2017/publish-zomato-medium.md