Most dashboards fail because they try to answer everything. A simple method to design screens around the decisions people make.

One question per dashboard

Decide the question before you pick the charts. Are we on track this month? Which branch needs help? A dashboard that tries to answer everything answers nothing well.

Lead with the answer

Put the headline number or status at the top, then the context that explains it. The reader should know in two seconds whether to worry.

Make the next action obvious

A good dashboard ends in a decision. Link each warning to the screen where it can be fixed.

  • Name the user and the decision
  • Pick the fewest charts that answer it
  • Show change and target, not only the total

Cut what nobody uses

Review the dashboard after a month. Anything nobody opens goes, and the rest gets room.