E-Commerce

Nova Commerce

Enterprise E-Commerce Platform

Industry: Retail & Consumer Brands
Duration: 5 months
Delivered: 2024
Client: Nova Commerce (demo)

Next.jsTypeScriptNode.jsPostgreSQLStripe-ready

Nova Commerce puts several brand storefronts on a single commerce core, so catalogue, pricing and promotions are managed once and published everywhere. Storefront pages render at the edge for fast loads, while the commerce API owns cart, checkout and promotion logic. Payments and fulfilment feed a reconciliation process that operations teams can audit line by line. Built as a demo, it shows how multi-brand retail can avoid the maintenance cost of separate shops while still giving each brand its own merchandising surface.

The challenge

What the business was up against

Each brand had grown its own storefront, its own checkout and its own admin, and the seams showed in both customer experience and operations. Products were duplicated across systems with drifting prices, promotions had to be recreated by hand for every shop, and finance reconciled payments from several different exports. Peak traffic exposed how tightly checkout was coupled to page rendering, so a slow storefront could stall orders. Small merchandising changes required developer releases because nothing was configurable, and adding a new brand meant cloning a platform rather than launching a storefront. The result was rising maintenance cost with falling release confidence.

At a glance

Industry: Retail & Consumer Brands
Duration: 5 months
Delivered: 2024
Engagement: Discovery → delivery → support

The solution

What we designed and shipped

We separated the storefront from a shared commerce core and gave each brand its own presentation layer. Catalogue, pricing and promotion rules live once in the commerce API and are consumed by every storefront, so a price change publishes everywhere at the same time. Checkout and payment flows sit behind a stable API, decoupling order capture from page rendering. Reconciliation jobs match payment captures to orders and surface exceptions for finance. A merchandising admin lets teams schedule campaigns without a deployment, and storefronts render through the edge cache to keep p95 load low under peak traffic.

At a glance

Primary category: E-Commerce
Services: E-Commerce · Web · Enterprise
Team: Product, design, engineering, QA
Delivery: Two-week increments

Key features

What the platform actually does

Product Management

Catalogue, variants and merchandising rules shared across brand storefronts.

Order Management

Order lifecycle from checkout through fulfilment and returns handling.

Customer Accounts

Profiles, addresses and order history across every brand storefront.

Payments & Reconciliation

Payment capture matching with exception queues for finance teams.

Analytics & Merchandising

Conversion, basket and campaign performance reported by storefront.

Multi-storefront Support

Brand-specific themes and domains running on one shared core.

Architecture

How the system is put together

Each tier can be scaled, replaced or taken offline independently. Data flows left to right; failure in a downstream tier never blocks the primary transaction path.

Storefront

Brand-specific presentation backed by shared commerce services.

SSR/ISR pagesProduct and category pages rendered and revalidated at the edge.
Edge cachingCached responses keep peak traffic off the origin core.

Commerce API

Stable contracts for cart, pricing and promotion behaviour.

CartSession cart with inventory checks and pricing rules.
PricingCurrency, tax and promotion calculation in one service.
PromotionsScheduled campaigns and discount rules applied consistently.

Operations

Post-purchase processes that keep money and stock aligned.

PaymentsAuthorisation, capture and refund handling per transaction.
FulfilmentOrder routing, dispatch confirmation and returns intake.
ReconciliationCapture-to-order matching with exception reporting for finance.

Data

Transactional storage, session cache and an analytics event feed.

PostgreSQLOrders, catalogue and customer records for the core.
Redis sessionsFast session and cart state for active shoppers.
Event stream to analyticsCommerce events streamed for near real-time reporting.

Technology stack

Chosen for the decade, not the demo

Every dependency here has a long support horizon, an active community and a large hiring pool. That keeps total cost of ownership predictable long after launch.

Demo project

Next.jsTypeScriptNode.jsPostgreSQLStripe-ready