The proliferation problem
Ask a large enterprise how many dashboards it maintains and the answer is usually an estimate. Ask how many are opened weekly and the answer is usually unknown. Both facts point at the same cause: reporting was commissioned faster than agreement on what mattered.
A metric needs three things
Before a number belongs on a dashboard, it should have an owner, a threshold and an action. Who watches it, at what value does it become interesting, and what happens then?
Metrics without all three are context. Context is useful, but it should not be presented with the same weight as a metric that triggers a decision. Most enterprise dashboards make no visual distinction between the two.
If nobody can say what they would do differently at a different value, it is not a KPI. It is a fact.
Lagging and leading
Executive reporting tends to concentrate on outcomes the organization can no longer influence. Revenue for a closed quarter is a result, not a lever. It belongs in reporting, but it does not belong in the operational review where decisions are made.
The harder work is identifying the leading indicators — the measures that move first and that a team can act on this week.
One definition per metric
Where two functions calculate the same measure differently, meetings turn into reconciliation. The resolution is organizational rather than technical: one agreed definition, one owner, and a documented method. The calculation is the easy part.
Retire what is not used
Reporting estates need pruning on a schedule. If a dashboard has not been opened in a quarter, it should have to justify its continued existence. This is unpopular and consistently worthwhile.