How We Built ElephantBird β A Full-Stack Fashion E-Commerce Platform for an Indian Retailer
From system design to production deployment β how IRDWorld built a fast, maintainable, and fully client-owned e-commerce platform for ElephantBird, an Indian fashion brand.
The Brief
ElephantBird is an Indian fashion brand selling original clothing designs directly to consumers. When they came to IRDWorld, they had a clear requirement: a production-grade e-commerce store that could handle a growing product catalogue, process orders reliably, and be managed independently by a non-technical team after handoff.
No vendor lock-in. Full code ownership. A store that works β and keeps working without us.
The Architecture Decision
The first decision was the stack. For a fashion e-commerce platform with SEO requirements, fast product page loads, and a non-technical admin, the stack needed to satisfy three things:
- βServer-rendered pages for SEO β product pages need to be indexed by Google
- βA managed data layer that the client could own without ongoing database maintenance
- βAn image pipeline that handles fashion photography at scale without manual processing
We chose Next.js, Supabase, Cloudinary, and Vercel. Here is why each decision was made.
Stack Breakdown
Next.js
Server-side rendering was non-negotiable for a retail store. Product pages need to be indexed by Google from day one. Next.js gave us SSR with React's component model β fast pages, clean separation between server and client components, and a deployment model that works natively with Vercel.
Supabase
Supabase gave us managed PostgreSQL with built-in authentication and row-level security β a production-grade data layer with zero ongoing maintenance overhead post-handoff. The client does not need to manage a database server. Row-level security meant we could scope admin access correctly without building a custom auth system.
Cloudinary
Fashion photography is high-resolution and high-volume. Cloudinary handles automatic WebP conversion, responsive image resizing, and CDN delivery. After initial upload, the client never touches image processing again. Every product image is automatically optimised and served from the right size for the right device.
Vercel
Native Next.js deployment with zero-config CI/CD. Every push to main triggers a production deploy. Every pull request gets a preview environment. Post-handoff, the client manages deployments through a dashboard β no terminal required.
What We Built
- βProduct catalogue with category browsing, filtering, and variant selection (size, colour)
- βFull cart and checkout flow β address collection, order summary, confirmation state
- βCloudinary image pipeline β automatic WebP, responsive sizes, zero manual processing post-handoff
- βSupabase auth and row-level security for correctly scoped admin access
- βAdmin panel for products, inventory, and order management β no code required
- βVercel pipeline with preview and production environments
- βFull technical documentation and clean codebase handover at project close
The Handoff
Every IRDWorld project ends with a structured handoff. For ElephantBird this meant full technical documentation covering the system architecture, database schema, deployment workflow, and admin panel usage guide. The client received the complete codebase, all credentials, and a walkthrough session.
Post-launch: zero issues. The store is actively trading. The client manages products, inventory, and orders independently.
Results
"IRDWorld delivered exactly what we needed β a robust, fast, and easy-to-manage store. The image pipeline and admin panel have made running the business significantly easier. Professional team, clean handoff, zero issues post-launch."
The live store is at elephantbird.in.
Building something similar?
Tell us what you need. We respond with a technical assessment and an honest proposal.
Start a Discussion β