Friday afternoon. Someone sends the file they have spent two days on and writes: finished. On Monday morning their manager sits in front of it and spends three hours reworking it. She says nothing, because she does not want to spoil anyone's Friday, and thinks quietly that next time she will just do it herself. Both of them did their job. Both of them meant something different by finished.
This is where a task turns into a dependency. The cause sits in a single word that the handover hangs on and that means two different things to two people.
Clean up the kitchen
Say the same sentence to four people: please clean up the kitchen. For the first it means the dishes are in the sink instead of on the table. For the second it means the dishwasher is running. The third includes the wiped counter, the fourth also the towel hung straight. All four did what you asked. All four did something different.
And if you then go through it again yourself, it looks to everyone as though nobody but you does it properly. The cause was in the sentence. It carried no picture of what a clean kitchen looks like.
Why "do your best" gets you nothing
Edwin Locke and Gary Latham spent thirty-five years studying how goals affect performance. More than a hundred different tasks, over forty thousand participants, eight countries. Their finding fits into one sentence in American Psychologist: when people are asked to do their best, they do not do so.
The reason sits right next to it and is the heart of this article. Do-your-best goals have no external referent, so each person defines them for themselves. Everyone sets the bar where it belongs from their point of view, and everyone stays inside what they understood. In those same studies, specific, difficult goals beat the call to do your best with effect sizes between 0.42 and 0.80.
What the vagueness costs a team
Since 1970 this has had a name and a measurement. John Rizzo, Robert House and Sidney Lirtzman developed the role ambiguity scale in Administrative Science Quarterly, for the uncertainty about what is expected of you and how good work is recognised.
Travis Tubré and Judith Collins pulled that research together in the Journal of Management. Role ambiguity relates negatively to job performance, with a correlation of around minus 0.21. The revealing part is the comparison inside the same paper: role conflict, meaning contradictory demands from two directions, comes in at only around minus 0.07. Contradictory instructions weigh less than the absence of any instruction about how finished work is recognised. People cope with a conflict they can name. What was never said, they work straight past.
The criterion belongs to the task
That also settles where the criterion lives. It hangs on the task and on the role doing it, not on the person currently doing it.
While "finished" sits in your head, you are the body that checks it, and every handover runs through you. Once it hangs on the task, someone else can take it over and see for themselves whether the result fits. It is the same mechanism as with the knowledge that lives in one head: what has been said out loud can be passed on.
The idea is old and well documented
How much an explicit criterion does shows up in a study with nothing to do with office work. Alex Haynes and colleagues introduced a nineteen-item list in eight hospitals, read aloud at three fixed points: before anaesthesia, before the incision, and before the team leaves the room. Inpatient complications fell from eleven to seven per cent, and deaths fell from 1.5 to 0.8 per cent. Published in the New England Journal of Medicine in 2009.
Every item on that list was already familiar to everyone involved. The only new thing was that they were said out loud, every time, in the same order.
In software development the same thing is called a Definition of Done. The Scrum Guide describes it as a formal description of the state in which a result meets the required quality measures, and draws a clear consequence: what does not meet the definition counts as not delivered and goes back. That is software vocabulary for something every team needs. A shared sentence about when something counts.
How to start this week
Take a task that regularly comes back to you because something was missing.
- Describe the finished result, not the activity. Three to five points an outsider can tick off. "Quote reviewed" is not one. "Price, delivery time and payment terms are in it" is one.
- Write down what it unlocks. What is the result needed for, what happens with it next? Whoever knows that gets the small decisions along the way right on their own.
- Name the role that checks. Which role looks at it before it goes out. Without that line you stay the checking body, even with the criterion written down.
- Put it where the task happens. In the template, in the ticket, in the list that gets opened anyway.
- Extend it on the next round. Whatever came back regardless was not yet in the criterion. That is exactly the next line.
One task a week. After a month the work that goes back and forth most often has a shared finished.
The rest is criteria
A "finished" that exists only in your head stays an expectation, and expectations cannot be handed over. Written out, it becomes something everyone on the team can see and reach, and the question of why this had to be reworked again comes up less often.
The next part of the Clarity Playbook looks at who actually decides. Each part brings a method that builds on the one before.
One last thought, in case this reads as controlling other people. Writing down how a finished result is recognised can look like distrust towards people who do their work well. In practice it is the condition for letting go. While the criterion sits in your head, you have to look at every result yourself. Once it is written, someone else can check, and the Monday mornings you used to spend reworking come back to you.
Care for your systems, and you care for your people.
Elisa
This article is written by Elisa Winkelmann with AI assistance and under her editorial review.
Sources
- Edwin A. Locke and Gary P. Latham, "Building a Practically Useful Theory of Goal Setting and Task Motivation: A 35-Year Odyssey", American Psychologist 57(9) (2002), pp. 705-717. On do-your-best goals having no external referent, and on the effect sizes of specific goals.
- John R. Rizzo, Robert J. House and Sidney I. Lirtzman, "Role Conflict and Ambiguity in Complex Organizations", Administrative Science Quarterly 15(2) (1970), pp. 150-163. Origin of the role ambiguity scale.
- Travis C. Tubré and Judith M. Collins, "Jackson and Schuler (1985) Revisited: A Meta-Analysis of the Relationships Between Role Ambiguity, Role Conflict, and Job Performance", Journal of Management 26(1) (2000), pp. 155-169. Source of the values of around minus 0.21 and minus 0.07.
- Alex B. Haynes and colleagues, "A Surgical Safety Checklist to Reduce Morbidity and Mortality in a Global Population", New England Journal of Medicine 360(5) (2009), pp. 491-499. Eight hospitals, nineteen items, inpatient complications from 11.0 to 7.0 per cent and deaths from 1.5 to 0.8 per cent.
- Ken Schwaber and Jeff Sutherland, The Scrum Guide (November 2020). On the Definition of Done and what follows when a result does not meet it.