EatExpress
Dual-platform food delivery for UAE — 261 screens across a complete consumer ordering experience (75 screens) and a full restaurant partner operations platform (186 screens), unified under a single token-based design system.
My Role
Principal Product Designer
Duration
4 months
Year
2023
Tools
Figma · FigJam · Lexend Deca
Project Brief
Design a market-adapted, feature-advanced food delivery platform for UAE/Dubai — both a consumer ordering app and a full restaurant partner operations platform — going beyond the Zomato/Swiggy reference brief with UAE-specific features: AED, Arabic, +971, halal dietary filters, and Split Bill.
Dual-platform food delivery for UAE — 261 screens across a complete consumer ordering experience (75 screens) and a full restaurant partner operations platform (186 screens), unified under a single token-based design system.
Platform Architecture
2 apps. 261 screens. 1 brand system.
Consumer ordering experience and a comprehensive restaurant operations management platform — distinct visual identities, shared design system.
Customer App
Food delivery consumers — UAE market
Complete ordering journey: discovery → cart → checkout → tracking → Split Bill → Diet Mode
Partner App
Restaurant owners and operators
Full restaurant operations: orders, inventory, analytics, ads, invoicing, dual-track feedback, GSTIN
My Ownership Breakdown
The Problem
Zomato in Dubai. But Dubai isn't India. The brief said 'replicate Zomato.' The product required market-adapted original thinking.
The UAE food delivery market is distinct: AED currency, Arabic-first interfaces, +971 phone numbers, UAE payment methods, halal dietary requirements, and a restaurant partner landscape with GSTIN-equivalent compliance needs. A straight Zomato port would have been wrong for this market.
No UAE-market-adapted reference
Existing apps in UAE (Deliveroo, Talabat) have different interaction models. A Zomato clone wouldn't be competitive in this market.
Restaurant partners had no analytics
Partners needed outlet-level analytics, menu-to-cart conversion data, and multi-location comparison — not just order notifications.
No Split Bill for group dining
Group dining is a primary use case in Dubai's restaurant culture. Contact picker + equal/custom split was absent from all reference apps.
No Arabic RTL architecture
Supporting Arabic required RTL-aware component design from the start — not a post-design localisation layer.
UX Audit
What was broken before we redesigned
A heuristic evaluation of the existing system identified these critical issues before any design decisions were made.
Zomato/Swiggy not designed for UAE
Indian-market UX patterns (INR, domestic payment rails, Hindi content) don't map to AED, Arabic, or UAE food culture.
Market FitPartner analytics were minimal
Reference apps showed order counts. Operators needed outlet-level conversion analytics and trend indicators.
Analytics GapNo financial compliance UI
GSTIN, invoice recovery, bank account management — critical for UAE restaurant operator compliance, absent from consumer-focused apps.
Compliance GapOur approach: Use the brief as a floor, not a ceiling
Delivered the full Zomato/Swiggy feature set as baseline — then added Split Bill, Diet Mode, outlet-level analytics, dual-feedback management (Complaints vs. Reviews), GSTIN compliance, invoice recovery, and 6-language RTL support. Two distinct visual identities (red Consumer, orange Partner) under one token system.
Process & Approach
How I approached the challenge
Competitive Research
Benchmarked Zomato, Swiggy, Deliveroo, and Talabat. Identified 12 features absent from the reference brief that UAE market conditions required.
Dual-Brand Design System
Lexend Deca type system (35+ named styles), dual-brand tokens (red Consumer, orange Partner), 6-step scales with AAA accessibility ratings.
Customer App (75 screens)
Home → Restaurant → Cart → Checkout → Live Tracking → Split Bill → Diet Mode → Wallet. AED, +971, Arabic, halal filters, Dubai neighborhoods.
Partner App (186 screens)
Order pipeline (Preparing/Ready/Picked), Inventory, Outlet Analytics (conversion rate), Ads, Dual-track Feedback (Complaints vs Reviews), GSTIN, Invoicing, Support.
Design Principles
Two products. One DNA. Different personas.
The consumer app and partner app share every token but serve fundamentally different mental models.
Market-adapted, not market-copied
The brief said Zomato. The product required UAE-native thinking: AED, Arabic, +971, UAE payment methods, halal filters, Dubai location context.
Dual-brand from one token system
Red for the consumer (appetite, warmth). Orange for the partner (energy, business). Same spacing, typography, component patterns. Different identity, same DNA.
Partner app is a SaaS product
186 screens covering the complete restaurant lifecycle isn't a companion app — it's a SaaS operator platform. Analytics, compliance, and invoicing are first-class features.
Empty and error states first
Diet Mode with no matching restaurants. No orders pending. Network error during checkout. Both apps designed with empty states before happy paths.
261 screens where the 200th is as well-considered as the 1st — that's the standard.
Why This Was Hard
Complexity signals
What set this project apart from standard product design work — the technical, systemic, and domain-specific challenges that demanded real depth.
Two full apps — two design disciplines
261 screens across 2 apps with separate visual identities. Consumer UX and operator SaaS UX are fundamentally different disciplines applied simultaneously.
Split Bill — group dining fintech
Contact picker, equal/custom split amounts, payment confirmation per participant — a mini-fintech feature absent from all reference apps.
Diet Mode — dietary overlay system
A toggle that filters all restaurant listings and menu items simultaneously — a systematic filter architecture, not just tag labels.
Outlet-level conversion analytics
Restaurant partners see menu-to-cart conversion rate, outlet comparison, and trend indicators (↑/↓) — a business intelligence layer on top of order management.
Dual-track feedback management
Complaints and Reviews as separate modules — different lifecycle states (Open/Dismissed/Resolved/Expired) and different operator workflows.
6-language + RTL architecture
Arabic (default UAE), English, French, Italian, Greek, Japanese — RTL-aware component architecture built into the system from the start.
261 screens. Two apps. UAE market adaptation. Split Bill. Diet Mode. Outlet analytics. One designer.
UI Screens
Customer App + Partner App
261 screens — complete ordering journey and full restaurant operations platform.
Feature Highlights
Features beyond the brief
Split Bill
Contact picker → equal/custom split → payment confirmation per participant. Group dining fintech absent from all reference apps.
Consumer InnovationDiet Mode
Toggle overlay filtering all restaurant listings and menu items simultaneously by dietary preference — systematic filter architecture.
Consumer InnovationPartner Order Pipeline
Preparing → Ready → Picked Kanban with per-order countdown timers, issue reporting, and delivery partner status.
Operations UX6-Language + RTL
Arabic (default), English, French, Italian, Greek, Japanese — RTL-aware component architecture built from the start.
LocalisationDesign System
Tokens, typography & components
A shared token-based system built to ensure consistency across every screen and state.
Color Tokens
Type Scale
Components
User Flow
Consumer ordering journey
Home Feed
Diet Mode + wallet
Restaurant
Menu + cart
Checkout
AED + promo
Live Track
Map + timeline
Split Bill
Group dining
Tech Stack
Design tools
Design
My Contribution
What I owned
01
Principal Product Designer
Complete design ownership: competitive research, dual-brand design system (Lexend Deca, 30+ color tokens), 261 screens across 2 apps, interactive Figma prototypes.
Impact & Performance
Project scale
261
Total screens
75
Customer App
186
Partner App
35+
Type styles
Timeline
4 months: brief to dual-app prototype
Month 1
Research + Design System
Competitive benchmarking. Lexend Deca type system, dual-brand token architecture, component library.
Month 2
Customer App (75 screens)
Full consumer journey — home, discovery, cart, checkout, Split Bill, Diet Mode, wallet, 6-language, UAE adaptations.
Month 3–4
Partner App (186 screens)
Order pipeline, inventory, outlet analytics, ads, dual-track feedback, GSTIN compliance, invoicing, support.
Outcomes
The impact we shipped
Measured results from usability testing, stakeholder reviews, and post-launch adoption tracking.
75
Customer App screens
186
Partner App screens
261
Total screens
35+
Named type styles
"Delivered a market-differentiated dual-platform food delivery application with 261 screens — going significantly beyond the reference brief with UAE-specific features, original product thinking, and a production-ready design system."
Learnings & Reflection
What I'd do differently
01
Audit the reference before accepting it
Treating 'replicate Zomato' as the product ceiling would have produced a poor-fit product for UAE. Market research before brief acceptance — always.
02
Partner-side UX is its own problem
186 partner screens took as long as 75 consumer screens. GSTIN, invoice recovery, and outlet analytics are each as complex as any consumer feature.
Interested in working together?
Let's talk about your project or open role.
Live Prototype