By Andreas Pettersson, Founder, Leaders ADAPT
The Golem effect is the reason a manager who has quietly written someone off is almost always proved right within a quarter. The expectation leaks, the person reads it, the performance follows, and the manager ends up with evidence for a conclusion she reached first.
This page is the effect in 40 words, the 1982 study, three ways low expectations leak into a 1:1, the Ringelmann effect that gets searched alongside it, and the 1:1 structure that stops the leak.
I run Leaders ADAPT and wrote the 5-Minute Leader. The 1:1 Protocol in it is built so the manager's behavior stays the same whether or not she believes in the person that week, which turns out to be the whole cure.
Quick answer: The Golem effect is a self fulfilling prophecy in which a manager's low expectations lower an employee's performance. Babad, Inbar and Rosenthal showed it in 1982, as the mirror of the Pygmalion effect. It leaks through behavior: cancelled 1:1s, level three tasks only, answers before questions. The fix is a 1:1 that runs the same way every week regardless of the manager's private view.
The Golem effect, defined
The Golem effect is a self fulfilling prophecy in which low expectations held by a supervisor or teacher lead to poorer performance by the person they are held about, through changes in how that person is treated. The person does not fail on their own. They are helped to.
The study, in three lines
Robert Rosenthal and Lenore Jacobson showed in 1968 that teachers told certain pupils were about to bloom got better results from those pupils, who had been picked at random. In 1982 Elisha Babad, Jacinto Inbar and Rosenthal showed the mirror image: teachers' low expectations, especially among teachers prone to bias, produced worse performance from the pupils they held them about. They named it after the golem, the clay creature of Jewish folklore that does what it is made to do.
The effect does not need a word to be said. It travels through tone, time and the tasks handed over.
Three ways it leaks into a 1:1
| The leak | What the manager does | What the report learns |
|---|---|---|
| Time | Cancels this person's 1:1 first when the week is busy | I am the lowest priority |
| Tasks | Hands over level three only, do exactly this, check in | I am not trusted to decide |
| Talk | Answers before the question is finished, skims the update, never asks what they recommend | My thinking is not wanted |
Each of these is a reasonable response to a report the manager believes is weak. Each of them produces a weaker report. Within a quarter the manager has her evidence, and the person has a reputation that will follow them to the next job.
Boomerang check: does your work stay delegated, or does it come back?
Three questions, scored to the protocol that fixes the leak. The 5-Minute Leader is the four protocols, Delegation, 1:1, Accountability and Rhythm, written down so a manager can run them in about five minutes a day, so tasks stop boomeranging back half done.
Get the 5-Minute Leader or see what is inside the 5-Minute Leader.
The Ringelmann effect, since you will see it next to this one
The Ringelmann effect is different and gets confused with the Golem effect because both describe performance dropping for reasons unrelated to ability. Max Ringelmann, a French agricultural engineer, measured rope pulling around 1913 and found that force per person fell as the group grew. That is a group size problem, the original measurement behind social loafing, and the social loafing page covers it.
The Golem effect is a 1:1 problem. One person's view of another. The two meet in one place: the report the manager has written off tends to be given group work, where the low expectation is confirmed by loafing nobody can see. Two effects, one person, one quarter.
The fix: a 1:1 that does not depend on what you believe this week
Telling a manager to expect more of someone does not work, for the same reason telling a Theory X manager to trust people does not. The belief has evidence. What works is a structure that holds the manager's behavior constant, so the expectation cannot leak, and the evidence has a chance to change.
The 1:1 Protocol I teach does this with four fixed moves.
The meeting is never the one that gets cancelled. Same slot, every week, every report. Reliability is not rationed by the manager's opinion.
The report goes first. 40/40/20. The manager opens with "What's on your mind?" and says nothing for the first forty percent. The report the manager has written off is the one who most needs to be heard before being told.
One task a week goes over at level one. Decide and tell me. The authority level is stated out loud, because the Golem leak is strongest in the tasks, and a person given only level three work will behave like a person who can only do level three work.
Feedback is specific and checkable. SBED: the situation, the behavior a camera would have seen, the effect, the change. Not "you need to be more strategic", which is the expectation leaking as an adjective. A specific demand can be met, and meeting it is the evidence that changes the manager's mind.
The one on one meetings page has the full structure, and coaching employees is the version for the manager who wants to move someone from written off to backed.
The honest exception
Sometimes the low expectation is correct, and the person is in the wrong role. The structure above is still the right move, because it produces clean evidence either way. Three months of weekly 1:1s, level one tasks and specific feedback will show a person who rose to it or a person who did not, and in both cases the manager knows something she only believed before. The clear expectations page is where that evidence gets written down.
The 1:1 that keeps expectations from leaking
The Golem effect travels through the manager's behavior, and the 1:1 Protocol in The 5-Minute Leader fixes the behavior in place. I wrote the book after scaling a company to more than 150 people and realising, late, that two people I had written off had been managed into it by me. Four protocols, Delegation, 1:1, Accountability and Rhythm, each run in about five minutes a day, so every report gets the same meeting, the same handover and the same quality of feedback regardless of what the manager privately thinks that week.
The 1:1 Protocol holds the 40/40/20 structure, the opening question, the energy question, SBED and the 24/48 Rule. The Delegation Protocol holds the three authority levels, with the rule that every person on the team gets level one work, not just the favourites.
There is a question in the 1:1 Protocol, asked in the second forty percent, that reverses the Golem effect faster than anything else I have found, and it is four words long.
Golem effect FAQ
What is the Golem effect?
The Golem effect is a self fulfilling prophecy in which low expectations held by a manager, teacher or other authority lead to lower performance from the person they are held about. The manager expects little, behaves accordingly, the person reads it and delivers little, and the manager is proved right. It is the dark twin of the Pygmalion effect.
What is the difference between the Golem effect and the Pygmalion effect?
Direction. The Pygmalion effect is high expectations producing better performance, shown by Rosenthal and Jacobson in 1968 with teachers told that random pupils were about to bloom. The Golem effect is low expectations producing worse performance, shown by Babad, Inbar and Rosenthal in 1982. Same mechanism, opposite sign, and the Golem side is the one managers do by accident.
What is an example of the Golem effect at work?
A manager decides a report is "not strategic". She stops asking his opinion, gives him level three tasks only, cancels his 1:1 when busy and skims his updates. Within a quarter he stops offering opinions, waits to be told, and takes no initiative. She tells HR he was never strategic. The expectation produced the evidence.
How do you avoid the Golem effect as a manager?
Behave as though the expectation is open, even when your private view is not. Keep the 1:1 every week, let them go first, ask what they recommend before you say what you think, hand over at least one task at level one, and give feedback in a specific, checkable format. Expectations leak through behavior, so change the behavior first.
What is the Ringelmann effect?
The Ringelmann effect is the finding, from Max Ringelmann's rope pulling experiments around 1913, that output per person falls as the size of a group grows. It is the original measurement behind social loafing. People search for it alongside the Golem effect because both describe performance falling for reasons that have nothing to do with ability.
Is the Golem effect the same as the Ringelmann effect?
No. The Golem effect is about expectations: one person's low view of another lowers that person's performance. The Ringelmann effect is about group size: each person's effort falls as more people share the task. One is a 1:1 problem, the other a team design problem. Both are fixed by making the individual visible and expected to perform.
Name the person you have written off
There is one. Everybody has one. This week, do not cancel their 1:1, let them go first, hand them one level one task and give them one piece of feedback with a situation and a behavior in it. The Golem effect runs on your behavior, and your behavior is the one thing on this page you control.
Which of your reports gets the level three work, and when did you decide that?


