Multi-Brand Product Discovery Storefronts

Challenge

The client operates several consumer shopping brands that each need a fast, modern storefront with product discovery, search, comparison and favourites. Building and maintaining a separate codebase per brand — and per market — would multiply engineering effort and make every improvement a job repeated five times over. They needed a single storefront platform that could be deployed under different brands and languages, while sharing one set of features, one AI search backend and one deployment process.

Solution

We built a shared Next.js storefront template that powers multiple brands from one codebase, including Octer (UK), Octer DE (German market) and Fevuh. Each deployment is the same application configured for a different brand, locale and product category tree, so a feature built once ships to every brand. The storefronts connect to the Cur8or AI engine for semantic product discovery and to a GraphQL content API for editorial and merchandising, giving shoppers search, product comparison and favourites backed by AI relevance. To make running many brands practical, we built a deployment automation layer that generates per-brand configuration and ships each storefront to Google App Engine without manual setup.

Project Results

A single platform now serves multiple brands and markets, including a fully localised German edition, with shared features and a shared release process. New brands can be launched by configuration rather than by rebuilding, and improvements roll out across every storefront at once. The deployment automation lets the same codebase run as isolated per-brand instances across the UK and EU.

Features and Benefits

• One Codebase, Many Brands: Each brand is a configuration of the same app, so features are built once and shared everywhere.
• AI Product Discovery: Cur8or-powered semantic search surfaces relevant products beyond simple keyword matching.
• Compare & Favourites: Shoppers can shortlist and compare products for a smoother path to purchase.
• Full Localisation: The German edition shows the platform scaling cleanly to new languages and markets.
• Automated Multi-Tenant Deployment: Per-brand config generation ships isolated instances to the cloud with no manual setup.

Technologies Used

• Next.js & React: Server-rendered storefront for speed and SEO across all brands.
• Tailwind CSS: Shared, themeable styling system for consistent yet brand-distinct design.
• Cur8or REST API: AI semantic search, category trees and product data.
• GraphQL (content API): Editorial content, merchandising and store data.
• Google App Engine + deployment automation: Per-brand environment generation and scalable hosting.

Development Process

We designed the template brand-first: everything brand-specific — copy, theme, category tree, locale — lives in configuration, while the application logic stays shared. The storefront was wired to the Cur8or engine for AI search and to a GraphQL API for content, so each brand gets the same powerful discovery experience. We proved the localisation model by shipping a full German edition of Octer alongside the UK site. To keep operating many brands sustainable, we built automation that generates each brand's deployment config and pushes it to Google App Engine, turning "launch another brand" into a configuration task rather than an engineering project.