Perception is not reality, but it should not be ignored
A perception does not capture reality like a camera. It does, however, record how someone interprets what they could observe within a particular relationship, at a particular moment, and against specific expectations. That limitation does not make perception irrelevant; it determines how perception should be read.
In talent management, we tend to swing between two extremes. The first treats a manager’s or group’s opinion as a direct description of the person. The second dismisses every perception as subjective and seeks evidence supposedly free of interpretation. Neither approach reflects how information is produced within an organization.
Perception also says something about the relationship
Someone who works with a person every day has access to behaviors that a distant manager does not see. Someone who receives their work observes effects that may not appear in aggregate results. The person knows about decisions, constraints, and attempts that others can only infer. Each source has different access and different blind spots.
A difference between peer and manager ratings could reflect inconsistency in the person’s performance, different expectations, different kinds of interaction, or unequal access to evidence. Attributing it immediately to bias wastes information; averaging it away may conceal it.
Research on agreement between self-ratings and ratings by others in leadership shows that the issue resists a simple interpretation. The review by Fleenor and colleagues found varied metrics, factors affecting both the people being rated and the raters, and inconsistent findings (Fleenor et al., 2010 (opens in a new tab)). Disagreement exists, but its meaning depends on how it arose.
A perception needs a defined focus
Perception becomes especially fragile when the question is broad. Are they a good communicator? Do they show leadership? Are they strategic? Such questions invite different episodes to be compressed into a single overall impression. Reputation, visibility, and verbal fluency can take the place of evidence.
The situation changes when the focus is defined through observable behaviors. For example: explains a decision so that the people carrying it out can reconstruct the purpose, constraints, and next step. The request is no longer to judge an abstract quality, but to provide relevant situations, contexts, and effects.
That does not eliminate interpretation. Two people can still disagree about the clarity of an explanation or the complexity of the context.
The pattern matters more than the vote
When several sources are available, an average may appear neutral. But three different ratings are not necessarily three interchangeable measurements of the same phenomenon. A peer, a direct report, and a client may each have observed legitimately different facets.
Reading the pattern requires keeping at least four elements visible: source, context, behavior, and period. If the rating changes with the relationship, that variation may be part of the phenomenon. If it changes within the same relationship without an apparent change in context, examine occasion-specific noise, memory, or criteria. If every source agrees but each knew only a single episode, unanimity does not compensate for sparse evidence.
Assessment standards remind us that validity applies to the interpretation of a result for a defined use, not to a score in the abstract. They also require warnings when a common interpretation lacks support and consideration of unintended consequences (AERA, APA, and NCME, 2014 (opens in a new tab)). In an organizational context, this means asking which conclusions a set of perceptions actually supports and which decision it is intended to inform.
Evidence always comes with a perspective
Documents, results, and work samples may seem more objective because they exist outside the evaluator. Even so, someone decides which result matters, which period to compare, which part of the work to attribute to one person, and which standard to apply. Evidence reduces some ambiguity; it does not eliminate the need for judgment.
The alternative to a weak perception is therefore not indiscriminate data collection. It is a disciplined combination of sources whose relevance to the subject can be explained. A deliverable shows a result; an observation provides behavior and context; the person’s explanation reveals decisions or constraints; an additional source helps assess stability. Each piece serves a different purpose.
What a perception can reveal
A perception can show how a behavior was received, what effect it had on a relationship, or where expectations are misaligned. It can draw attention to a pattern that warrants investigation. It can reveal that the same behavior works under some conditions and fails under others.
On its own, it cannot define the full extent of someone’s capability or explain its cause. Nor should it be dismissed simply because another source tells a different story. Its value emerges when it stops competing to be the definitive version and instead occupies a traceable place within a broader interpretation.
The decision improves when it preserves that dual truth: every perception is partial, and some perspectives reveal exactly the information that needed to be seen.
Keep exploring
Sources of Power: when expert intuition deserves a place in assessment
Sources of Power makes it possible to treat expert intuition as situated recognition without turning experience or confidence into unquestionable authority.
How to add a 360 without turning it into an average
A 360 improves an assessment when it preserves what each source observed, from which relationship, and within what limits; averaging first destroys that information.
How to design a calibration session that does not turn into a negotiation over scores
How do you keep a calibration session from turning into a negotiation over scores? Bring evidence, context, and criteria into the conversation before the number.
Confidence is not competence
Confidence becomes visible before competence does. It appears in the speed of a response, the firmness of a voice, and the willingness to take up space. Competence, by contrast, can be supported only by relevant evidence of what a person can do under specific conditions.
Why leaders distrust skills assessments
When a leader distrusts a skills assessment, the usual reaction is to attribute it to resistance to change. But distrust may also point to genuine problems with burden, observation, interpretation, or consequences that the design has not yet resolved.
The leader who assesses without observing
The authority to assess comes with the role. The quality of the judgment comes from the evidence. When the two are confused, familiarity, reputation, and visible results take the place of observation.