halo effect
Also called: halo error
A rating error in which one strong impression, good or bad, colors the judgment of everything else about a person. Rating each criterion separately against recorded facts reduces it.
The halo effect is a rating error in which one strong impression of a person, good or bad, spills over into judgments that have nothing to do with it. A confident speaker gets rated as more competent; someone who saved a crisis gets rated higher on teamwork and quality too; one visible mistake drags every other rating down (sometimes called the reverse halo). It is one of the oldest and best-replicated findings about ratings: scores on different qualities of the same person agree with each other far more than the person's actual behavior does.
For a lead it shows up wherever people are judged: performance reviews, promotions, hiring, and quieter choices such as who gets the interesting work or the benefit of the doubt. Visible strengths cast the longest shadow: speed, confidence, ease in meetings, a dramatic save. Steady, quiet contributions get the least light.
Knowing about the halo effect doesn't switch it off, so the fix is in the process. Write the criteria before you look at anyone's results. Rate each criterion separately, on recorded facts, ideally one criterion across all people before moving to the next, rather than one person across all criteria. Keep a fact log through the whole period, and ask a second reader to challenge any adjective that has no fact behind it.
Example
Review season at Klasna, August 2027. In July a payment-page fault flooded the queue and Taras carried it; the CEO thanked him in the company chat. The lead's first draft reads: "Taras was our star this summer and is the best coordinator on the team." Rated one criterion at a time on January–July facts, the picture splits:
The July week is real and belongs in the review as a fact: he handled a flood well. The halo is letting it answer questions it says nothing about. A fairer line: "Highest volume on the team and strongest under a sudden spike (week of 12 July); reopen rate 9.07% against a team rate of 5.32% is the main area to work on next year."
Common mistakes
- Thinking awareness is enough. Knowing the bias doesn't remove it; change how you rate.
- Rating one person across all criteria at once. Rate one criterion for everyone, then the next.
- Letting one event answer every question. A crisis save says a lot about pressure and little about everyday quality.
- Writing adjectives without facts. "Star", "careless", "a natural" are signs the halo is writing for you.
- Missing the reverse halo. Quiet people and one visible mistake get marked down across the board.
Learn it in the course
- Evaluating people fairly · Keep a fact log, spot halo, recency and similar-to-me bias, use the same criteria for everyone and ask for a second opinion.