Why your dashboard never matches finance.

Where the gap actually comes from.

Next time, name the reasons instead of apologizing. Almost every marketing-versus-finance delta is some mix of these six:

Gross versus net. Your dashboard counts order values; finance counts revenue after tax, shipping and discounts.
Returns and cancellations. Finance books them when they happen; your dashboard never heard of them.
Test and fraud orders. Filtered in one system, counted in the other.
Timezones. A sale at 23:40 lands on different days in different tools.
Attribution windows. The platform claims a conversion days after the click; finance booked it on the purchase date.
Modeled data. GA4 fills consent gaps with estimates; your ERP does not estimate anything.

None of these is wrong on its own. What is wrong is that nobody decided which definition wins where.

The fix is a decision, not another dashboard.

The definitions have to live in exactly one place, and that place is the data warehouse, not the heads of five analysts. We take the raw GA4 export in BigQuery and model it in layers with Dataform, so every definition is written down once and reviewed like code:

What to demand from your data team.

You do not need to write SQL to steer this. Ask for four things: a definitions document a non-analyst can read; automated checks that alert when yesterday’s numbers drift outside tolerance against the source system; one reporting table that marketing dashboards and finance exports both read; and changes by reviewed request, so a definition never silently shifts under your KPIs.

When those four exist, the Monday number matches, and the meeting is about what to do next instead of whose export is right.