do you actually trust your analytics numbers?

building in public here, launching tomorrow.

honest question for anyone who ships product analytics: when your dashboard says signups are down 9%, do you believe it, or do you go check in the database?

i never believed it. not because the tools are wrong exactly, but because i could never see how they got there. and now every analytics tool has an AI layer that will happily generate a query and hand you a confident number with no way to check it.

so the thing i built answers in plain english but only ever runs deterministic reports, and there is a CI test that fails the build if the dashboard, the API and the MCP tools ever disagree with each other.

curious whether this is a real problem for other people or just my own paranoia. what do you currently do when a number looks wrong?