Case study

Multi-Channel Commerce Integration and Automation Platform

A SaaS platform that connects wholesale brands to 40+ commerce, accounting, inventory and EDI systems. It has processed 4.1 million orders, with a vendor app, a retailer app and an admin console.

SaaS product engineeringIntegrationsEDIBilling and subscriptionsTest automation
Order syncWeyBeeWeyBee
Web storeMarketplaceEDI
Integration hub
ERPAccounting3PL
4.1Morders processed
40+platform integrations

The client

Wholesale brands sell through online stores, marketplaces, retailers and sales agencies, while running inventory and accounting in yet more systems. Each channel sends orders in its own format, including EDI from large retailers. The client's product connects all of these into one operating system for the brand, so orders, stock, prices and invoices move automatically.

The challenge

  • Connect 40+ very different platforms, from store builders and marketplaces to accounting, inventory and ERP systems and EDI trading partners.
  • Let vendors control data flow themselves: mappings, filters, swaps and schedules, without writing code.
  • Stage incoming wholesale orders so teams can review, fix and release them before fulfilment.
  • Run a subscription business on top: plans, per-order billing, coupons and referral payouts.

What we did

Our engineers build and run the platform: .NET 10 APIs for the vendor and admin applications, a set of background workers for order sync, price updates, vendor data, analytics, per-order billing and the retailer directory, and two React and TypeScript front ends. A visual Sync Builder lets vendors design their own data flows. Each vendor has its own database, so tenants stay isolated. Browser automation tests cover key flows.

What it does

Integration and sync engine

Install and authorise integrations, then control data flow with a visual Sync Builder, field mappings, filters and schedules; sync status and event logs show exactly where a record failed.

Order and invoice processing

Stage, edit and release wholesale orders and invoices; match customers; pick, pack and ship; print packing slips; inspect raw EDI.

Catalogue and inventory

Products and variants, on-hand stock allocated across channels, and scheduled price updates.

Wholesale network

Vendors connect with retailers and invite sales agencies; partner orders flow straight into the vendor's pipeline.

Sales analytics

Sales by month, day, channel and agency, top sellers, order status and estimated savings from automation.

CRM and commissions

Customers, sales reps, leads, CRM activity and commission tracking.

Retailer app

Retailers accept vendor invitations and receive their network orders in the same application.

Admin console

Integration catalogue and sync rules, EDI chains and document maps, account management, billing, a harvested retailer and agency directory, and customisation requests.

How it is built

  • .NET 10 APIs with JWT authentication, EF Core and Dapper; background workers on the .NET Generic Host.
  • Single-tenant dataA central platform database plus a separate database for each vendor.
  • Vendor app on React 18 with a React Flow sync builder; admin app on React 19 with Vite.
  • EDI document mapping, PDF and barcode generation, and an accounting desktop connector.
  • Stripe and Chargebee billing; product analytics and onboarding tools.

Results

  • 4.1 million orders processed through the platform.
  • 200+ separate apps connected into one operating system for brands.
  • 100+ vendors and retailers running automated order flows.
AI Engineering · Next

AI-ready by design.
Where AI fits next.

4.1 million orders and 40+ integrations give this platform rich, structured data. AI can make the data flows smarter and the vendors faster.

AI field mapping

Suggest field mappings when a vendor connects a new system.

Order exception triage

Explain failed syncs in plain language and propose the fix.

Sales insight assistant

Let brands ask questions about sales by channel, agency and product.

Start with an AI use-case

One to two weeks to pick the use case and prove the data is ready.

Add AI to a system like this
DATAAlready in the platform
AI ENGINEModel to add
INSIGHTSignals surfaced
PRODUCTInside the workflow
ACTIONDecisions automated
Ask AiBee