A Client Asked for a Dashboard, What They Actually Needed Was a Process

The request was clean: build a single dashboard pulling together paid, email, and product data so leadership could stop asking five different teams for five different numbers every Monday. A reasonable ask, a well-defined deliverable, a clear budget. I nearly just started building it.

A few discovery conversations in, I asked what decisions the dashboard was supposed to support. The honest answer, eventually, was: none specifically. Different leaders wanted different numbers because different leaders were optimizing for different things, and nobody had actually agreed on what the team was trying to move. The dashboard request wasn't a reporting problem. It was a symptom of unresolved disagreement about priorities, dressed up as a tooling gap.

What building the dashboard as requested would have done

A unified dashboard built on top of unresolved metric disagreements doesn't resolve the disagreement — it just gives everyone a shared, authoritative-looking surface to keep arguing on. I've seen this exact pattern before: the dashboard ships, two VPs immediately disagree about what a number on it means, and the dashboard gets blamed for being 'wrong' when the actual problem is that it faithfully reflects a definitional conflict nobody wanted to have out loud.

A dashboard can't resolve a disagreement about what matters. It can only make that disagreement louder and better-formatted.

What we actually did instead

• Ran a short alignment workshop first, forcing the leadership team to agree on a small set of metrics that actually mapped to decisions they made
• Documented metric definitions explicitly, since half the disagreement turned out to be semantic — two teams using 'active user' to mean different things
• Built the dashboard only after that alignment existed, as a presentation layer on an agreement that already existed rather than a substitute for one
• Scoped a lightweight monthly review process alongside the dashboard, since a dashboard with nobody accountable for acting on it decays into background noise within a quarter

The dashboard that eventually shipped was simpler than the one originally requested — fewer metrics, fewer views. It also actually got used, because it reflected decisions people had already agreed to make, instead of trying to manufacture agreement after the fact.