Data

From dashboards to decisions: building a metrics layer

Omar Haddad, Head of AI & DataJun 20266 min read

Two dashboards, two definitions of 'active user', two different numbers in the same meeting. Without a shared source of truth, analytics stops being evidence and starts being ammunition.

A metrics layer fixes this by defining each metric once — the exact events, filters and grain — in version-controlled code that every tool reads from. 'Active user' means one thing, everywhere, because it's defined in one place.

Model the events cleanly upstream first: consistent names, a stable identity, and the handful of properties you'll actually slice by. A metrics layer built on messy events just gives you consistent nonsense.

With definitions centralised, self-serve becomes safe. Analysts and PMs can explore without quietly reinventing a metric, and a change to a definition updates every dashboard at once instead of drifting apart over months.

The goal isn't more charts — it's fewer arguments. When everyone trusts that the number means what it says, the conversation moves from 'is this right?' to 'what do we do about it?'

Let's build what's next.

Tell us about your product and we'll come back within one business day with a clear path forward — no obligation.

Headquartered in Hong Kong — engineering and delivery across Eastern Asia and Eastern Europe, with US partners as fractional CTOs.

Start a project

Tell us what you're building — no obligation.