Perhaps you stopped offering someone complex work after a difficult first project. Perhaps you trust a dependable colleague so much that you barely look at what has changed. Before your next decision, ask yourself: what is my view based on now?
You may have good reasons to keep it. The point is to give new information a fair hearing, including information that makes a positive impression less comfortable.
Our suggestion: choose one judgment that affects a real decision. Separate what happened from your explanation, check what you might be missing, then decide whether anything needs to change. You do not need to reassess everyone after every 1:1.
Separate the event from your explanation
“The afternoon team got the handover information late twice” describes something you can check. “Lea is unreliable” is a judgment about Lea. Before it influences the next assignment, check what the team had agreed and when the information arrived.
Was the deadline clear? Did the person have the inputs and authority they needed? Have there been comparable handovers that went well? These questions need not excuse a missed deadline. They help you identify what needs addressing: an individual action, the way work is organized, or both.
Research finding: selective interpretation.
When you already have a view of someone, it can shape which examples you look for and how you read them. Nickerson (1998) reviews this tendency, known as confirmation bias. The review describes a general tendency. Whether it applies here depends on the evidence in this case.
Our suggestion: look for a concrete counterexample as well as confirming examples. Check both with the same care. An old note is a starting point, not proof that its interpretation still holds.
Ask what could change your view
Before the conversation, try: “What would I expect to see if my explanation were incomplete?” If the answer is “nothing could change my mind”, examine whether you are discussing a specific problem or defending a label.
Research finding: a laboratory strategy.
In the second experiment by Lord and colleagues (1984), 30 students chose questions to ask a stranger. Those given a profile describing the opposite personality showed less preference for questions that fit their starting assumption than the comparison group did. The study did not test managers’ 1:1s.
Our suggestion: The researchers supplied the profile. Coming up with another explanation yourself before a 1:1 is our adaptation. “The afternoon team got what they needed late twice. Can we go through what happened? I may be missing part of the story.”
Give the person time to check a detail or come back to it later. Compare your own recollection and theirs with the same evidence. The time an email arrived might help you both. A colleague’s opinion of the person’s character tells you little about that delay. Stick to the work. If the person brings up a private matter and wants support, discuss where they can get it.
One situation can lead to different conclusions
Imagine Lea on the early shift in a customer-service team. Twice, the afternoon team got the handover information after the agreed time. Her manager is starting to wonder whether Lea can be relied on and is thinking of asking someone else to coordinate the next shift. Lea says another team sent her the information late.
In this fictional example, Lea and her manager check when the information reached each team and who was expected to flag a delay. What they find changes what the manager does next.
The explanation changes
The records show that the required information arrived after Lea’s deadline.
Was the agreed handover time workable?
The manager revisits the assignment decision and coordinates a feasible handover with both teams.
The concern remains specific
The information arrived on time; the handover was late and no warning was sent.
What got in the way, and what should happen when a delay becomes likely?
Lea and her manager discuss what held up the handover and how to flag a delay next time.
Suppose neither team can confirm the timing yet. If the assignment can wait, leave it open until you can check. If someone needs to coordinate today’s shift, explain the temporary choice to Lea. “The duty lead is already covering today’s shift, so I’ll ask them to coordinate it. Tomorrow I’ll check the handover times with you before I decide who takes the next shift.”
A single successful handover would not necessarily resolve a repeated problem. Equally, one mistake does not establish a permanent trait. Consider comparable work and the significance of the decision. Reallocating a routine task and deciding a promotion require different levels of scrutiny and the appropriate company process.
Act on what you learn
Research finding: change can go either way.
Heslin and VandeWalle (2008) describe studies in which managers watched recorded negotiations. Managers who saw more room for people to change were more likely to recognize both improvements and declines in performance. They did not observe managers in 1:1s with their staff.
Our suggestion: keep both directions open. A trusted colleague may need support with unfamiliar work; someone who struggled earlier may now be ready. Neither outcome follows simply from believing people can change.
If you got it wrong, say what you got wrong and which decision you’ll revisit. “I thought the delay happened on your shift. Now I can see the information reached you late. I was wrong to put that on you. I’ll look again at who gets the next assignment.” Only promise what you have the authority to do.
If the concern remains, explain the specific evidence and expectation, hear any disagreement and agree what can be done next. You can acknowledge an upstream problem and still address an avoidable delay in flagging it. If nothing material has changed, keep a justified decision and explain why. A recurring work problem may need a different arrangement, not another request to try harder.
Suppose you trust a colleague to check customer quotes. After a pricing change, you spot an old rate on a quote they checked. Ask which price list they had. If the shared list was out of date, correct it. If the new rule was unclear, offer to check the next quote together. If the list was current and your colleague knew the rule but missed the old rate, say which price slipped through. You and your colleague could spot-check a few upcoming quotes, then decide together whether a second check is still useful.
Formal performance, conduct, discrimination, harassment, health or legal matters belong in company procedures, with HR or specialist support as needed. This reflection does not replace those processes.
A short note before the decision
Our suggestion: use only the lines that help. Avoid permanent labels, speculation about private life or unnecessary details. Private notes still need proper handling; see what to write after a 1:1.
Decision I need to make:
What happened, when, and how I know:
My current explanation, and what else could explain it:
Relevant information still to check, including the person’s account:
What would change or strengthen my view:
Decision or next check, with a reason and an update date if needed:
If the gap is that you have stopped learning about someone’s work, start with how you give people attention. The preparation checklist can help you bring a specific question to the next conversation.
Sources
- Raymond S. Nickerson (1998). Confirmation Bias: A Ubiquitous Phenomenon in Many Guises. Review of General Psychology, 2(2), 175–220. Repository copy.
- Charles G. Lord, Mark R. Lepper, Elizabeth Preston (1984). Considering the Opposite: A Corrective Strategy for Social Judgment. Journal of Personality and Social Psychology, 47(6), 1231–1243. Author copy.
- Peter A. Heslin, Don VandeWalle (2008). Managers’ Implicit Assumptions About Personnel. Current Directions in Psychological Science, 17(3), 219–223. Author copy.