Your platforms can talk, but does your business actually work as one?
Integration vs connectivity: why "the systems talk" is the wrong test, and what a governed event contract actually proves.
Making two systems talk is a week of work: a webhook, a nightly sync, a point-to-point call. Making a business run as one, across everything it depends on, is a different problem, and it’s the one most integration projects quietly leave unsolved.
”Integrated” is a low bar
Integration is a technical fact: data moves from A to B. You can demo it by Friday. What a demo doesn’t test is whether both sides agree on what the data means once it lands. Point-to-point integrations rarely share a canonical definition of “customer” or “order”: three systems, three slightly different truths, patched over by a nightly reconciliation job.
The mechanism: a contract, not just a pipe
Connectivity starts with a decision most integrations skip: which system is the system of record for each entity, and what the event contract leaving it promises. Get that right and consumers can be built idempotent, safe to replay without double-processing. Skip it and every new integration adds another undocumented special case to a web nobody can reason about six months later.
Every new expectation (faster service, AI reading live data) lands on this layer first, and a governed contract is the difference between adding a consumer in an afternoon and a three-month project to renegotiate what a field means everywhere it’s read. Two systems that talk isn’t proof of anything. A contract that survives a new consumer and a schema change is.