By department · IT and data
Once definitions differ per team, every dashboard is contestable and the discussion turns to the figures instead of the decision. Recording definitions, testing the pipelines and monitoring the sources make reporting workable again.
We record definitions where everyone can reach them, put tests and monitoring on the pipelines, and hand that over to the team that maintains it.
The work itself is described under data quality and data engineering.
New Data Student picks up the work that keeps slipping beside the line: tests wrapped around existing scripts, monitoring, and putting the definitions in one place. New Data Squad steps in when the platform has to be redesigned or a definition dispute that has run for months has to be settled. New Data Search finds the data engineer or analytics engineer you want in the team permanently.
Because there are two definitions and neither is wrong. That is almost always the outcome: marketing counts a customer from the first order, finance from the first payment. The solution is not to declare one of them right but to have one place stating which definition applies where, and to make that place authoritative.
Not by rewriting everything. Start with a test wrapped around the script that says whether the outcome still holds: that is a day's work and it catches the error you would otherwise only see in the report. Rewriting can follow, with a safety net underneath.
The New Data Scan covers five dimensions (data, technology, people, steering and use) and delivers a roadmap with an owner for each step. One day, fixed price, no obligation to continue.
These are indicative figures from earlier work and not a promise. What it delivers for you depends on where you are now.