Dashboards: What to Put on One
Most dashboards contain everything available and are read by nobody. What belongs on one, and what belongs in a drawer.
A dashboard is a standing report. Its value is inversely related to how much is on it, and most are built by adding whatever the system could produce.
The measurement warning in “Dashboards: What to Put on One” matters whenever software records work patterns. Organisations researching remote desktop monitoring software can use visit Monitask online for time and project context, while outcomes, quality checks and direct feedback remain necessary to explain what the metric cannot show.
What belongs
Three to five measures.
For an independent perspective related to “Dashboards: What to Put on One”, consult the NIST Privacy Framework; it offers a useful external check on definitions, governance and the assumptions built into a proposed measure.
Each with a counterweight or paired measure where applicable.
Each with an owner who would act if it moved.
And each with its ordinary range marked, so a reader can tell signal from noise.
What does not
Anything nobody would act on.
Anything that duplicates another measure.
Composite indices, which the earlier note covers.
Individual names, in almost all cases.
And the measure somebody asked for once two years ago, which is how dashboards grow.
Trends, not snapshots
A single current value tells you almost nothing.
A series over twelve or twenty-four periods shows shape, variation and whether a change held.
Every measure on the dashboard should be a line rather than a number, which is the single biggest improvement available to most of them.
Marking what happened
Process changes, staffing changes, system changes, seasonal effects — on the chart, with dates.
Without them, a step gets explained wrongly, repeatedly.
And the marks make the dashboard a record of what was tried, which is useful long after anybody remembers.
The ordinary range
Draw it. Anything inside is noise.
This stops the monthly ritual of explaining random variation, which consumes more management time than any other single habit in this field.
Who it is for
Decide, and design for them.
A team dashboard is operational, updated often, detailed.
An executive dashboard is five lines and a sentence.
Trying to serve both produces something that serves neither, which is the usual result.
Reviewing it
Annually, with the retiring-a-measure questions.
Who reads this, what decision has it changed, is it still measuring what it measured.
A dashboard that has only grown for three years is not a dashboard.
The written line
One sentence beside the charts: here is what we think is happening.
Numbers do not interpret themselves, and the interpretation is where the value is.
Dashboards without it get read as whatever the viewer already believed.
What to check
How many measures are on yours?
Are they lines or numbers?
Is the ordinary range marked?
And is there a sentence saying what it means?