Product update
Reading a competency report: distributions, drilldowns and exports
An organisation-wide competency chart is a summary of hundreds of judgements. How to read one without over-reading it — and what the latest release changes for HR admins.
By the CLEAR Talent team8 min read
Key takeaways
- An average competency score hides the shape of the result. Look at the distribution before you look at the number.
- A band is a grouping, not a verdict. Know where its boundaries sit and what the people near them look like.
- Drill down before you act: from the organisation to the group, to the person, to the assessment the result came from.
- An export that loses its context — period, scope, filters — becomes a different document. Make sure the context travels with it.
Once a review cycle closes, the competency data stops being about individuals and starts being about the organisation. HR is asked where the capability gaps are, which teams need investment and whether last year’s development spend moved anything. The answer usually starts with a chart — and a chart of competency results is a summary of hundreds of separate judgements, each made by a particular person, in a particular assessment, against a particular role.
The latest CLEAR Talent release is about the path from that chart back to the judgements underneath it. Admin competency reporting and its drilldowns have been improved; the demonstration bands that group results have been refined; the distribution and grouped report visuals have been restored; results now carry the context of the assessment they came from; and detailed exports have been finalised. Notifications are also now handed off between performance review stages. All of it is live. This article covers what changed and, more usefully, how to read a competency report so that it supports a decision rather than replacing one.
What an organisation-wide competency chart hides
The most common competency report is also the least informative: an average level per competency, across the organisation or a department. It is easy to produce and easy to present, and it answers almost nothing. A department averaging three out of five on stakeholder management could be a team where everyone is solidly competent, or a team where half the people are strong and half are struggling. Those two teams need completely different responses, and the average describes them identically.
The distribution is the first thing to read. How many people sit at each level, and does the shape make sense for the role? A cluster at the top of the scale for a competency that is new to most of the team is worth questioning. So is a flat spread for a competency the whole team uses every day, which often says more about how differently managers applied the scale than about the team itself.
The second thing to read is who is in the population. A report run across every role mixes people who are expected to be at level two with people expected to be at level four, and the result describes nobody. Competencies are mapped to roles, with target levels per role, so the meaningful question is usually how people compare with what their own role requires — not with each other.
An average tells you where the middle is. The distribution tells you whether there is a middle at all.
Bands are groupings, not verdicts
Most competency reporting groups results into bands — broadly, how fully a competency has been demonstrated against what is expected. Bands make a large dataset readable at a glance. They also hide everything inside each band, and they put a hard line between people whose results may be almost identical.
Two habits help. First, know where the boundaries sit and why. A band boundary is a design decision, and if nobody in the room can say what separates one band from the next, the chart is being read on trust. Second, look at the people near a boundary before acting on the band totals. A team where most people sit just below the line is in a very different position from one where most sit well below it, even though the chart shows the same count.
This release refines the competency demonstration bands and the drilldowns beneath them, and restores the competency distribution and grouped report visuals. The intent is that a band can be opened rather than taken at face value: the view should get a reader from the grouping to the results inside it without leaving the report.
Drill down before you decide
A pattern in a competency report is a reason to look, not a finding. Before it turns into a training budget or a talent decision, follow it down.
From the organisation to the group
Which teams, departments or roles produce the pattern? An organisation-wide gap is often a gap in two or three places, and the response belongs there.
From the group to the people
Is the gap shared, or concentrated in a few people? A shared gap suggests something structural — the expectation, the training, the opportunity to practise. A concentrated one is a conversation for those people’s managers.
From the person to the assessment
Which assessment did the result come from, when, and against which role? A result read without its assessment context can describe someone as they were two roles ago.
From the assessment to the evidence
Ratings in CLEAR Talent can require evidence. Where the report is about to drive a decision, check that the evidence behind the results supports the conclusion being drawn.
Two changes in this release sit on this path. Admin competency reporting and its drilldown have been improved, and assessment context results and drilldowns have been completed — so a result can be read alongside the assessment it came from, rather than as a free-floating number.
Drilling down also tells you what not to do. If a gap turns out to sit with people who have had no opportunity to use the competency, the answer is an assignment, not a course. And if it turns out to sit with one manager’s ratings, the answer is a calibration conversation, not a development plan.
An export without its context is a different document
Competency reports rarely stay in the platform. They are exported for a leadership pack, an audit committee, a workforce plan or a funding case, and from that point they are read by people who never saw the screen they came from. What travels with the numbers decides whether those readers understand them.
At a minimum, an exported competency report should say which period it covers, which population it includes, which framework and roles it was measured against, and which filters were applied. Without that, a chart of one department’s results can be read as the organisation’s, and a snapshot from before calibration can be read as the final position. Nothing about the numbers is wrong; the document has simply lost the information needed to read them.
CLEAR Talent already exports individual, team and organisation reports to PDF or Excel. This release finalises the admin competency context and detailed exports, and restores the competency report visualisations and the context that goes out with an export — so what leaves the platform says what it is.
Exports are also one of the actions the platform records against the user who made them. Competency data about named people deserves the same care outside the system as inside it: share the level of detail the audience needs, and prefer a summary where identifiable results add nothing to the decision.
Keeping a review cycle moving between stages
Competency reporting is only as complete as the cycle that feeds it. Each assessment period runs through stages — typically self-assessment, assessor review, calibration and final rating — and the place a cycle stalls is almost always the gap between two of them. One person finishes their part; the next person does not find out; the record sits waiting while the deadline approaches.
CLEAR Talent already sends automated reminders on whatever is outstanding in a stage. This release fixes how notifications are handed off between performance review stages, so that when one stage is done, the people responsible for the next one are told it is their turn.
The release also tightens the validation of KPI suggestions where some of their fields are empty, so an incomplete suggestion is handled cleanly rather than getting in the way of the goal it was meant to help with.
Neither fix changes how a cycle is designed. Both remove a reason for the cycle to slow down — and a cycle that finishes on time produces a competency report that describes the whole population, rather than whoever happened to complete first.
What to check before you present a competency report
Ten minutes with the report before it goes in front of leaders saves a much longer conversation afterwards.
Read the distribution, not just the average
For each competency you plan to highlight, look at how results are spread across the levels. If the shape surprises you, find out why before presenting it.
Check the population
Confirm which roles, teams and period the report covers, and whether people are being compared with their own role’s target levels.
Open the bands you are about to quote
Look at who sits near the band boundaries. A headline count can rest on several people who are one rating away from the next band.
Drill into at least one pattern
Pick the finding most likely to drive a decision and follow it down to the assessments and evidence behind it.
Check the cycle is complete
If stages are still open, say so. A report drawn from a cycle that is half finished describes the people who finished first.
Make sure the export says what it is
Before sending, confirm the export states its period, scope and filters, and that its level of detail suits the audience reading it.
A better report does not make the decision for anyone. What it does is shorten the distance between a pattern on a chart and the judgements behind it — which is the distance where most poorly grounded people decisions are made.