Treating finance and customer analytics like the same workload is how you end up with a platform that serves neither well.
Quick primer on CAP: consistency (everyone sees the same data), availability (the system keeps responding), partition tolerance (it works when the network splits). You get two. The third becomes a trade-off you make, knowingly or not.
The mistake I keep seeing: teams pick one CAP posture for the whole platform. Then every workload inherits it. Finance gets the latency finance doesn’t want. Analytics gets the consistency guarantees it doesn’t need. Nobody’s happy, and the platform bill keeps climbing.
Finance reporting is mostly C+A. Stale numbers in a board pack are a career-ending event, so availability without consistency is worthless here. A recommendation engine is fine with A+P and eventual consistency - a few seconds of drift doesn’t change anyone’s life.
So stop architecting one CAP answer. Architect per workload.
One line per data product. “Finance: consistency first. Analytics: eventual consistency is fine. Ops sync: pick partition tolerance over availability.” Pin it in the architecture doc. Revisit when the workload changes.
Most “our platform feels wrong” conversations are really workloads sharing a CAP posture that only fits half of them.
Can you write one sentence per data product describing which CAP properties it prioritizes?