All resources
Analytics18 September 2025·5 min read

Dashboards nobody opens: designing BI for the three decisions that matter

Most BI tools accumulate dashboards the way closets accumulate clothes — until nobody remembers what half of them are for, and nobody opens any of them anymore.

Ask most operations leaders how many dashboards their team has, and the number is embarrassingly high. Ask how many they actually open weekly, and it's usually two or three. The rest were built for a specific question months ago, answered that question once, and never got retired — cluttering the tool, diluting attention, and making the two or three dashboards that matter harder to find.

The three-decisions framework

Before building a dashboard, name the specific, recurring decision it exists to inform, and who makes that decision. If nobody can name one, the dashboard is a nice-to-have that will get built, viewed twice, and forgotten. Most teams, examined honestly, make only three to five decisions on a weekly cadence that genuinely need a live data view — everything else can be a monthly report or an ad hoc query.

What this looks like in practice

  • Interview the actual decision-makers about what number they check before making a specific call — not what metrics sound impressive in a review meeting.
  • Build exactly one dashboard per recurring decision, with the metric that drives the decision front and center, not buried among a dozen others.
  • Retire or archive dashboards nobody's opened in the last quarter — a shrinking, curated set of tools is more valuable than a growing, ignored one.

A useful test for any existing dashboard: if it disappeared tomorrow, would anyone notice within a week? If the honest answer is no, it was never serving a decision — it was serving a moment, and that moment passed.

The cost of dashboard sprawl

Every unused dashboard is a small tax on trust in the whole BI system — when half the tools are stale or irrelevant, users start doubting the ones that matter too, and default back to gut feeling exactly where data should have been driving the call.

Related capability

This is what our Predictive Business Intelligence practice is built around.

See how it's scoped