Dashboards nobody opens

Most organisations have a graveyard of dashboards. Commissioned with enthusiasm, built with care, opened twice, and never again. The data is accurate. The charts are clean. And still nobody looks.

The problem is almost never craftsmanship. It's that a dashboard is only as good as the decision it changes — and most dashboards aren't attached to a decision at all.

Why they go unused

• No decision behind them. They answer "what happened?" but not "what should I do?"
• Too much of everything. Forty metrics on one screen is a data dump, not a decision aid.
• No point of view. A number with no target or threshold is trivia. Is 4.2% good?
• Wrong cadence. A daily dashboard for a decision made quarterly — or the reverse.

If you can't name the decision a dashboard supports, you haven't built a tool. You've built wallpaper.

Design for the decision, not the data

Start from the question

Before a single chart, write the sentence the dashboard exists to complete: "When I see ___, I will ___." If no one can finish it, the dashboard has no reason to exist.

One primary metric

Lead with the number that matters, framed against a target. Everything else is context that explains it, arranged in order of how often it's actually needed.

Give every number a point of view

Show the target, the trend and the threshold that should trigger action. "4.2% — below the 5% goal, down from 4.8% last month" tells a story. "4.2%" doesn't.

Make it a habit

Usage is a design problem, not a training problem.

• Match the cadence to the decision. Deliver the view when the decision is genuinely made.
• Push, don't wait to be pulled. A short alert when a threshold is crossed beats a dashboard someone has to remember to open.
• Add a sentence of narrative. The one-line "here's what changed and why" is often the most-read part of any report.

The test that matters

A dashboard succeeds when someone changes their mind because of it. That's the only metric worth tracking about your metrics. Build for that, and the graveyard stops growing.

Tell us which decision you're trying to support and we'll design backwards from it.