Measuring Work That Has No Unit
Analysis, design, advice and management cannot be counted. Four approaches that work without a comparable unit.
Where the unit problem makes counting meaningless, measurement is still possible. It just measures something other than quantity.
The practical lesson in “Measuring Work That Has No Unit” is to connect every number to a decision and retain the context behind it. Teams exploring how to measure employee productivity can review a practical guide to measuring employee productivity as one source of operational evidence, provided the purpose is disclosed and the interpretation is tested with the people affected.
One: flow
How long things take from request to delivery.
For an independent perspective related to “Measuring Work That Has No Unit”, consult the Harvard Business Review productivity collection; it offers a useful external check on definitions, governance and the assumptions built into a proposed measure.
Needs no comparable unit, as the flow note argues.
Tells you about queues, waiting and whether the function is getting slower.
The single most transferable technique for professional work.
Two: rework
How often something comes back.
Also unit-independent, and it points at causes upstream — unclear briefs, missing information, handoffs.
In advisory and design work this is frequently the most informative measure available.
Three: commitment reliability
Of the things this function said it would deliver this period, how many arrived.
Measures predictability rather than volume.
Gameable by committing to less, which is why it is reported alongside what was committed, and that pairing works.
Four: asking
Structured, periodically: what took longer than it should have, what got in the way, what did you not get to.
Qualitative, specific, actionable, and the only method that reaches the causes.
Treated as soft and consistently more useful than any ratio, which is the honest finding from organisations that do both.
What these have in common
They measure the system rather than the people.
They produce actions rather than rankings.
And none requires deciding that one report equals another report, which is the fiction that counting demands.
What to avoid
Counting deliverables, which rewards splitting.
Hours, which the denominator note covers.
Self-assessed complexity scores used for assessment, which corrupt immediately.
And composite indices, which combine several of the above into something unreadable.
Choosing between them
If the complaint is slowness: flow.
If it is quality: rework.
If it is unpredictability: commitment reliability.
If nobody can say what the complaint is: asking, first.
The reporting that works
Two or three of these, with a short written account of what changed and why.
The written part is not padding — in professional work it carries most of the information, and a number alone invites the wrong reading.
What to check
Which of the four do you use for your professional functions?
Are you counting deliverables anywhere?
Has anybody asked the team what got in the way?
And does your reporting include a written account alongside the numbers?