It can, and the deciding factor is what people bring into the room. Calibration run on evidence raises the credibility of a decision. Calibration run on opinion launders it.
For the CHRO, HR leader or business leader who owns the performance system. Article 5 of seven in the performance management investigation.
Eight managers, a distribution on the screen, two hours. By the end the room agrees. Whether the people being discussed were judged more accurately is a different question, and it is rarely asked.
Three things that get treated as one
| What it looks like | How you would check it | |
|---|---|---|
| Agreement | The room settles on the same word | Two managers rate the same evidence separately |
| Accuracy | The word matches the work | Read the evidence with the name removed |
| Fairness | The person can see why | Say the reason out loud to them |
A room can reach complete agreement about someone nobody in it has seen work. That is consistency. It is not correctness, and it is not fairness to the person.
What decides whether it works
As practitioners we have observed that the value of calibration is set before anyone speaks, by what managers are expected to bring.
A manager who arrives with three examples, the context of the work, and what changed after feedback, is contributing evidence. A manager who arrives with "she is just very senior in her thinking" is contributing an impression, and impressions are where confident people beat careful ones.
That is a cultural question more than a procedural one. Does your organization treat the meeting as a place to justify a judgment, or as a place to negotiate an outcome? Are managers told beforehand that they will be asked for evidence, and does anyone follow up when they cannot produce it?
There is a line of research worth knowing about here. The effect of holding raters accountable, of having to justify a judgment to someone, has been studied directly since the 1990s 4, alongside the finding that training managers against shared standards improves accuracy 3.
What managers should be asked to bring
If preparation decides the outcome, it should be specified rather than hoped for. What we ask for, and what we have seen work as practitioners:
- Two or three examples per person, with the situation, what they did, and what followed. Not adjectives.
- What changed after feedback during the year, which separates a person who course-corrected from one who was never told.
- The context of the work, because a good year on an easy account and a hard year on a failing one look identical in a number.
- The evidence for the top and bottom calls specifically. Middle ratings rarely get argued; the ends do, and the ends carry the consequences.
- One sentence you would say to the person if they asked why.
A meeting where two managers cannot produce this is not calibrating. It is averaging impressions, and it can favour people whose managers are more articulate.
The case for the meeting
Set against the fashion for dismantling calibration, there is a real argument for it, and it is not only administrative.
A decision made by several people who have looked at evidence is more credible than a decision made by one person alone. It surfaces work the direct manager never saw. It exposes the manager who rates everyone highly and the one who rates nobody highly. And it signals something to employees that matters: that decisions about them are taken seriously enough to be examined by more than one person.
Field research supports part of this. Calibration meetings have been found to add information and to shift ratings in ways that relate to other performance measures 1. It also matters for a reason outside the meeting: how a decision is reached, and whether it can be explained, is the subject of a large literature on fair process 5.
And the case against, from the same literature
Research inside one company found the other half of the story 2. Managers shared information strategically, and adjustments fell unevenly across groups. The meeting is a negotiation as well as a review, and negotiations reward the confident, the senior and the well prepared, which is not the same list as the fair.
Both findings can be true in different rooms. Which one describes yours depends on preparation and culture, which is where we started.
Could you explain it to the person?
The most useful discipline in the room is to write, for each change made, the reason you would give the person affected.
"Your manager had not seen the work you did with the other team" is a reason. "We needed to fit the distribution" is a constraint, and if that is the honest answer, the organization should say so rather than dress it as a judgment about the person.
If a change cannot be explained to the person it describes, it has not been calibrated. It has been traded.
What the evidence supports, and how far
Calibration meetings can add information and shift ratings in ways related to other measures 1. Research inside one company found strategic sharing and uneven adjustments 2. Shared-standard training improves accuracy 3, and rater accountability has been studied as a distinct influence 4. Whether a decision can be explained, and how it was reached, has its own large literature 5.
The claim that preparation and culture decide the outcome is ours, consistent with these findings rather than demonstrated by them. So is the list of what managers should bring.
Evidence: contested.
Our position
Keep calibration, and spend the effort on what managers bring into the room rather than on the scale. Require evidence for the top and bottom calls, and write the reason you would give the person.
Take it into the room: the Before the calibration meeting worksheet turns this article into one printed page of questions, a framework and a table to fill in.
What would change this conclusion
Evidence that decisions coming out of calibration are better explained and hold up better when challenged, rather than simply more consistent with one another. Or a comparison of meetings run with required evidence against meetings run without it, which is the claim this article actually makes. Evidence that outcomes track who argues most effectively, after controlling for the work, would weaken the case for the meeting as most organizations run it.
Read next: What actually happens when organizations remove ratings?
Evidence map
- Speer, Tenbrink and Schwendeman (2019), Let's Talk It Out: The Effects of Calibration Meetings on Performance Ratings. Two field studies; author-posted text accessed, including the stated limitations. Association, not a randomized fairness test.
- Bol, De Aguiar and Lill (2025), Calibration in the Performance Evaluation Process. One-company study; publisher abstract accessed.
- Roch, Woehr, Mishra and Kieszczynska (2012), Rater Training Revisited. Meta-analysis; publisher abstract accessed.
- Mero and Motowidlo (1995), Effects of rater accountability on the accuracy and the favorability of performance ratings. Journal of Applied Psychology. Record and title verified through Crossref and Semantic Scholar; abstract not retrievable through the sources checked. Cited for the line of inquiry, not an effect size.
- Colquitt, Conlon, Wesson and colleagues (2001), Justice at the millennium: A meta-analytic review of 25 years of organizational justice research, and Colquitt, Scott, Rodell and colleagues (2013), Justice at the millennium, a decade later. Journal of Applied Psychology. Records and titles verified; abstracts not retrievable through the sources checked. Cited for the existence and scale of the fair-process literature.
Sources checked September 22, 2026. Prepared with AI assistance.