What to do when two managers see the same employee completely differently

It happens on almost every management team eventually. Two managers who've each worked with the same employee, maybe because the person changed teams, or the reporting line is shared, describe them in noticeably different terms. One sees someone reliable and strong. The other sees someone who needs constant follow-up. Neither manager is lying. They're both describing something real. They just aren't describing the same thing.

This isn't a rare edge case. Standard practice at most larger organizations includes a dedicated calibration process specifically because individual managers' assessments of the same employee routinely diverge, sometimes significantly, when compared side by side. Smaller teams without a formal calibration process don't have this problem less often. They just have no structured way to catch it before it causes real damage.

Why this happens even between fair-minded managers

A large part of the gap comes down to calibration standards that were never explicitly agreed on in the first place. One manager might consider a 5-out-of-5 the default unless something's clearly wrong. Another might reserve that same score for genuinely exceptional work, treating a 3 as the normal, solid baseline. Both managers can be completely consistent within their own standard and still land in totally different places when compared to each other.

Add in the fact that two managers often see different slices of the same person's work, one during a stressful reorg, one during a calmer stretch, one focused heavily on client-facing moments, another mostly on internal delivery, and the disagreement stops looking like a mistake and starts looking like the predictable result of two people with genuinely different, genuinely valid vantage points.

Disagreements are the point of calibration. If everyone already agreed, there'd be nothing to discuss.

Why the disagreement matters more than it seems

Left unresolved, this kind of gap doesn't stay contained to one conversation. Employees are highly sensitive to fairness, and unaddressed inconsistency in how people are evaluated is one of the more reliable ways to quietly damage morale and retention, even when no single manager did anything wrong. An employee who senses they're being judged completely differently depending on who's in the room, without understanding why, reasonably concludes the evaluation process itself isn't trustworthy.

The fix isn't picking a winner

The instinct when two managers disagree is to figure out who's right. That's usually the wrong question. A more useful approach redirects disagreement toward evidence rather than opinion: what specific outcomes support this view, and would we describe the same behavior the same way if it came from a different employee. The goal isn't to declare a winner. It's to check whether both perspectives are being applied against a shared, consistent standard.

Why this is so much harder without a written record

A disagreement based on two managers' unaided memories is nearly impossible to resolve productively. Neither person has anything concrete to point to beyond their own general impression, which means the conversation defaults to whoever argues more confidently, not whoever has better evidence. Pre-submitted evidence, specific accomplishments, dated incidents, concrete outcomes, is what allows a calibration conversation to be about facts instead of impressions.

This is exactly where a shared note history changes the conversation. If both managers have been logging dated, specific observations about this employee over the months they worked together, the disagreement stops being "I think they're solid" versus "I don't" and becomes something far more productive: two managers reading each other's actual notes and discovering, often, that they aren't really disagreeing about the employee at all. They're just each holding half the picture.

Keeping it from turning personal

These conversations work best when kept professional and specific to performance and behavior, not personalities, and confidential enough that people feel safe speaking candidly. A disagreement about how to interpret an employee's work is a completely different, much lower-stakes conversation than a disagreement that starts to feel like a personal conflict between two managers.

Grounding the conversation in specific, dated notes rather than general impressions naturally keeps it in that lower-stakes territory. It's much easier to discuss "here's what I observed on March 14th" without it feeling like an attack than it is to discuss a vague, unanchored sense that a colleague is wrong about someone.

The version of this that never needs a formal meeting

Large organizations solve this with scheduled calibration sessions. Most small teams don't need anything nearly that formal, as long as the underlying problem, two people forming an opinion from two different, unshared slices of the same relationship, gets solved earlier. If notes are visible across the whole management team as they're written, rather than only surfacing at review time, a diverging impression tends to get noticed and reconciled naturally, in the normal course of reading each other's notes, long before it becomes a disagreement that needs its own meeting.