Why we built Entity instead of reaching for a framework
Every framework we tried made the first month faster and every year after slower. The demo went up in a weekend; the invoice arrived over the following decade — in upgrade cycles, in plugins abandoned by their authors, in the slow discovery that the "batteries included" were the wrong batteries for a shop floor.
About ten years ago we made a different bet: own the whole stack, keep it small, and let it earn its complexity. That bet became Entity — the framework and development environment that now runs every major commercial solution we ship.
What owning the stack actually buys
It isn't pride. It's the ability to say yes. When a retailer asks for something their platform vendor considers unusual — offline tills that reconcile themselves, a product model with genuinely awkward variants, a checkout that changes behaviour by store — the answer doesn't depend on whether a plugin exists. The distance between "described" and "deployed" is our own code, all the way down.
- Deploys are instant, with zero downtime — because the deploy pipeline is ours too.
- Nothing ships without unit tests, and nothing has to work around a dependency we can't see into.
- A decade of production hardening in every new project — reliability compounds like interest.
The trade-off, honestly
Building your own foundation is slower at the start, and it demands discipline: every feature we add is a feature we maintain forever, so Entity has stayed deliberately small. It has no marketplace and no conference. What it has is millions of users a day passing through it without noticing — which is the only benchmark that has ever mattered to us.
If your platform has started sending you invoices measured in workarounds, we should talk.