Why dashboards go unopened
A dashboard gets built, gets attention for a fortnight, and then quietly stops being opened. Within a quarter the team is back in a spreadsheet that one person maintains by hand.
This is rarely a tooling problem. It happens because the numbers on screen don’t connect to a decision anyone actually makes, or because two people pulled the same metric, got different answers, and stopped trusting the whole thing.
Start from the decision
Before adding any metric, four questions are worth answering. They take a minute each and remove most of what would otherwise end up on the page.
- What decision does this number inform?
- Who makes that decision, by name?
- How often do they make it — weekly, monthly, quarterly?
- What would they do differently if the number moved significantly?
If the last question has no answer, you have found something interesting rather than something useful. Those numbers belong in a reference report people can go looking for, not on the dashboard that is supposed to drive a weekly conversation.
A metric nobody can act on isn’t a KPI. It’s a fact. Facts are fine, but they shouldn’t take up the top of the page.
What makes a metric usable
The metrics that survive their first year tend to share three properties, and the ones that get abandoned are usually missing at least one.
Owned
One named person is accountable for the number and for acting when it moves.
Without it a bad number is everyone’s problem and nobody’s job.
Defined once
One written definition and one source, so two teams cannot produce two answers.
Without it every meeting starts by arguing about the number.
Comparable
A baseline and a target, so anyone can tell good from bad without asking.
Without it the number is just a reading on a dial.
How many metrics is enough
Fewer than most teams expect, and the right number depends entirely on who is looking. A board wants a handful tied to outcomes. An operations team wants the ones they can influence this week. Those are different pages, not one page with everything on it.
The combined mega-dashboard is a reliable way to make a view nobody owns. Splitting by audience costs a little more to build and tends to be the difference between a dashboard that is used and one that is bookmarked and forgotten.
- Name the audience for the view.
- List the decisions that audience actually makes.
- Allow at most one headline metric per decision.
- Move everything else to a reference page, linked but not front and centre.
Write the definition down
Every KPI needs a short definition recorded somewhere people will find it. In practice this means six things: a plain-English description, the exact calculation, the source system, how often it refreshes, who owns it, and the current target.
Keep it next to the dashboard rather than in a governance document in a separate system. A definition nobody can find is the same as no definition, and it is what stands between you and the recurring conversation about why two reports disagree.
In short
Key takeaways
- Start from the decisions people make, not the data you happen to have.
- One owner, one written definition and one source per metric.
- Split dashboards by audience instead of combining them into one view.
