Telling Teams What Is Measured and Why
Measurement that is explained is tolerated and stays accurate. Measurement that is not gets managed instead.
People behave differently toward a measure they understand than toward one that simply appears. The explanation costs an hour and determines whether the figures stay honest.
The human context in “Telling Teams What Is Measured and Why” should be set before a workforce platform is configured. A team evaluating ethical employee monitoring built around trust for ethical employee monitoring can keep the rollout credible by stating the purpose, selecting only necessary settings and documenting how employees can review or correct records.
What to say
What is measured, precisely, including the definition of a unit.
For an independent perspective related to “Telling Teams What Is Measured and Why”, consult the ICO employment-practices guidance; it offers a useful external check on definitions, governance and the assumptions built into a proposed measure.
Why: what question it answers.
Who sees it and at what level.
What it will be used for — and what it will not.
And that it will be reviewed, with a date.
The sentence that matters most
"This informs how we improve the process. It does not inform how we assess people."
Said at the start and honoured, it keeps the data accurate because there is nothing to protect.
Said after somebody's figures were raised in an appraisal, it is not believed, and the data degrades from that point.
Explaining the counterweight
Tell them the pair: we count this, and we watch this alongside it, so that raising the first by damaging the second does not look like improvement.
This is reassuring rather than threatening, because it says the organisation knows the measure can be gamed and has thought about it.
Showing them their own data
Teams should see their own figures, first and routinely.
Hearing about your own numbers from somebody else's report is the fastest way to make a measure adversarial.
And teams spot errors in their own data immediately, which improves the quality of everything built on it.
Inviting the objection
Ask explicitly what the measure misses.
Write down the answers and publish them beside the measure as known limitations.
A measure with its limitations stated is used more carefully by everybody, including whoever reads it upward.
When it changes
Announce changes to definitions before they take effect, with the reason.
A definition that changes silently makes the whole series meaningless and the team will know before you do.
The review conversation
Quarterly or annually: is this still telling us anything, what would you change.
Short, and it is where most measurement improvement actually comes from.
Teams will suggest better measures than anybody outside the work would, given the chance.
What not to do
Introduce measurement with a message about performance expectations, which frames everything that follows.
Report team figures upward before the team has seen them.
Or describe a measure vaguely, which is read as concealment and produces exactly the defensive behaviour you were hoping to avoid.
What to check
Do your teams know what is measured and why?
Have they seen their own data before it went upward?
Are the known limitations published?
And has anybody been told what the measure will not be used for?