Composable commerce for USESI
Replacing a legacy ERP-bound storefront with a services-based platform that eight distinct brands could run on without any of them feeling like a tenant.
The storefront was bound to the ERP, which meant every commerce decision was really a decision about a system nobody wanted to touch. Eight brands shared the constraint and none shared the benefit: each had its own workarounds, its own spreadsheet of product data, its own reasons customers could not find what they came for.
The question underneath the replatform was not which vendor. It was which problems we should stop solving ourselves, and which problem is actually the business.
The technical design was the smaller half. The larger half was sequencing: which brand migrates first, which workflows earn a rebuild, which brands can share a catalog and which cannot because their customers use the same words to mean different products.
Getting eight of them onto one platform is stakeholder prioritization with an architecture diagram attached.
Each brand had its own P&L owner and its own reasons the shared architecture should bend its way. The roadmap had to make those trades visible instead of settling them in private.
We bought commerce capability and product-data management, then built and kept the integration tier: orchestration exposing legacy databases to the commerce platform and to external partners. Partner-facing endpoints have adopters, documentation, versioning pressure, and revenue attached, which makes them a product by any working definition.
A platform decision and a P&L conversation are the same conversation when the architecture is drawn around the money. Eight brands run on shared services, the product data is governed once, and the layer we refused to outsource is the one that pays.