Home / Blog / Note

"The assessment report a hiring manager will actually read"

"Most assessment centre reports rank candidates instead of showing evidence, so the hiring manager forwards a verdict nobody can check. A one-page format with three evidence bullets per dimension fixes that."

"The assessment report a hiring manager will actually read"
Representational image.

The PDF is fourteen pages. Page one is a table: nine competencies, a score out of five beside each, and an overall recommendation at the bottom in bold. Pages two to fourteen are prose. The hiring manager opens it on his phone, reads the table, forwards it to his own boss with the word “fine”, and books the offer letter. Nobody reads page seven, where the only sentence describing what the candidate actually did in the group exercise sits between two paragraphs about strategic orientation.

Six weeks into the job the candidate is struggling with exactly the thing page seven described. The report was not wrong. It was unreadable in the only sense that matters: the person paying for it could not get to the evidence before reaching the verdict, so he took the verdict.

Why the report drifts into a ranking

Three things push a report towards a scoreboard, and none of them is laziness.

The first is process. A centre ends with an integration session where assessors agree a rating per dimension, and that session produces a grid of numbers. When the report is written afterwards, the shortest path from grid to document is a table, and a table invites a rank. The scoring drift problem starts here: once numbers exist, the sentence explaining them feels optional.

The second is the buyer. Line managers ask for a recommendation, and so does the vendor’s own account manager, because a report without a verdict is harder to defend in a hiring committee. That is commercial pressure, not psychometric, and it changes what the document is for.

The third is the one nobody names. Assessment centre ratings tend to cluster by exercise rather than by dimension. Petrides and colleagues studied 487 individuals on nine dimensions, each measured across six exercises, and found strong exercise effects: the ratings grouped by the exercise a person had just done, not by the trait the exercise was meant to measure. The practical consequence is that a dimension score of four out of five is often one good hour in one room, repeated across a row of a spreadsheet.

An assessor writing fourteen pages of dimension prose is therefore asked to describe continuity the data does not show. The document gets longer exactly where it is thinnest.

What the manager is actually deciding

Start from the decision, not the person. A hiring manager at the end of an assessment centre is deciding three things, and only one of them needs the assessor.

  • Does this person clear the bar we set for this role, on each dimension we said mattered?
  • What would they do on day nine, when the script runs out and the room is difficult?
  • Is the gap between what we saw and what the role needs coachable in three months, or is it a different hire?

The first is the assessor’s. The second is the assessor’s too, because only they watched. The third belongs to the manager, who knows the team, the budget and the manager above him. A report that answers the third question has taken a decision it has no standing to take, using the least evidence in the pack.

The one-page format

One page, four blocks, in this order. If it does not fit, the centre did not produce enough evidence and the length is hiding that.

Block What goes in it Limit
Header Role, date, exercises run, who observed, decision deadline 5 lines
Bar check Each dimension as bar met, not met, or insufficient evidence one line each
Evidence Three bullets per dimension, each from a different exercise 25 words per bullet
Not seen What the centre could not test, and what the manager must probe in interview 3 bullets

Notice what is missing. No overall score, no ranking, no personality adjectives, no five-point scale. A scale invites averaging, and averaging is how a strong signal in one exercise becomes a three.

The “insufficient evidence” line earns the format its keep. It is the only honest option when a dimension was observed once, or when two assessors disagreed and the session resolved it by seniority.

Three bullets, each from a different exercise

The bullet is the unit of the whole document, so it has a shape: what the situation was, what the person did, what happened next. Under 25 words. One phrase in the candidate’s own words, because a manager recognises a person by how they talk, not by an adjective.

The weak version, which is what most reports contain:

Shows good influencing skills in the group discussion.

The version a manager can use:

In the tender meeting, the finance stakeholder had already said no. He asked what the no was protecting, then restated the client’s timeline in the stakeholder’s own terms. The stakeholder reopened the numbers.

Three tests for a bullet. Is there a verb the person performed, or only a judgement? Could another assessor in the room have written the same sentence? Could the manager ask the candidate about it on Monday without the candidate guessing where it came from?

The exercise-per-bullet rule does two jobs at once. It stops one strong room from defining a person, and it makes the report survive the question every good hiring manager asks: did you actually see this twice?

This is the same discipline as the contribution map for group exercises, applied to the write-up rather than the scoring. Airtime is not contribution, and a paragraph is not evidence.

Name the exercise that carried the rating

Because ratings cluster by exercise, the report should say which exercise produced the weight. One line under the dimension, in brackets: [client meeting, 2 of 3 bullets].

A reader can then judge the dimension for what it is. Two bullets from the client meeting and one from the written brief is a different claim from three bullets across three exercises, and the manager deserves to know which he is holding.

Where the centre ran a dimension only once, write it plainly: one observation, in one setting, on one day. Some buyers will push back. That pushback is information about the centre’s design, and it is better received in November than after a bad hire in March. If the design is thin, the fix is in the exercise matrix, not in more words.

The no-advice rule, and its one exception

The assessor does not write “recommended for hire”. The assessor writes bar met, bar not met, or insufficient evidence, with the evidence attached, and stops.

There is one exception, and it is narrow: when the manager asks a direct question in writing, the assessor may answer it as a hypothesis with the evidence attached. “If the role requires influencing a hostile stakeholder in month one, the tender meeting bullet is the closest thing we have, and it went well” is a legitimate sentence. “Strong hire, go ahead” is not, because it collapses three blocks of evidence into a number the manager cannot interrogate.

The rule also protects the assessor. A report that recommends becomes a report that can be blamed, and a report that can be blamed gets softer each cycle until it says nothing.

A worked example

Illustrative, with the numbers invented to show the shape.

A territory sales manager role, eight dimensions, five exercises: a client meeting, a group negotiation, a written brief, a role-play with an unhappy distributor, and a structured interview. One page.

Dimension Bar Evidence source Weight
Commercial reasoning Met written brief, distributor role-play, client meeting 3 exercises
Influencing without authority Met client meeting (2), group negotiation (1) 2 exercises
Coaching a junior Insufficient evidence interview only 1 setting, self-report

Under influencing without authority, three bullets: the tender meeting line above; a moment in the group negotiation where he gave away the pricing lever to keep the relationship and said why; and a silence he let run for eleven seconds before the buyer filled it. That third bullet is the one the hiring manager remembers, and it never appears in dimension prose because it is a behaviour, not a label.

The coaching dimension carries the insufficient evidence line and a note that the manager should run a 20-minute role-play in the final interview. That converts a gap into a task instead of a guess. If the manager later wants structured practice on giving that feedback, the coaching side of the house is the right referral, not another assessment.

Where this format stops working

One page is a forcing function, and forcing functions distort. Two honest costs.

First, the format rewards centres with a proper exercise matrix, where every dimension is observed at least twice in different settings, as the international guidelines for the method require. If your centre observes each dimension once, the one-pager will fill with insufficient evidence lines, and it will look worse than the fourteen-page version even though it is more accurate. Fix the matrix or accept the look.

Second, senior leadership assessment does not compress to a page, because the evidence is longer: a strategy presentation, a board simulation, a 90-minute interview. Use one page per group, three groups maximum, same bullet shape. The rule that survives is the bullet, not the page count.

A one-page report takes longer to write than a long one, because the writer has to choose. Budget half a day per candidate, and do not let the assessor who ran the wash-up write it cold at 9pm.

Monday

Take the last assessment report your team issued. Count the words per dimension and write the count in the margin. Then take the three dimensions that carried the most words and rewrite each as three bullets, each bullet from a different exercise, each with a verb the candidate performed.

If you cannot find three exercises, you have found the real problem, and it is not in the report. Send that finding to whoever designs the centre, and see whether the next cohort’s competency framework and exercise matrix are built to be reported on at all.

Sources

  • Petrides, K. V., Weinstein, Y., Chou, J., Furnham, A., & Swami, V. (2010). An investigation into assessment centre validity, fairness, and selection drivers. Australian Journal of Psychology, 62(4), 227-235. https://www.tandfonline.com/doi/full/10.1080/00049531003667380
  • International Taskforce on Assessment Center Guidelines (2015). Guidelines and Ethical Considerations for Assessment Center Operations, 6th edition. Journal of Management, 41(4). https://repository.up.ac.za/bitstreams/d677c925-9954-457b-ab0f-ac288e8d63d1/download
  • British Psychological Society, Psychological Testing Centre. Design, Implementation and Evaluation of Assessment and Development Centres: Best Practice Guidelines. https://www.klug-md.de/downloads/BPS_PTC_AC_Guidelines.pdf
  • Meiring, D., & Buckett, A. (2016). Best practice guidelines for the use of the assessment centre method in South Africa, 5th edition. SA Journal of Industrial Psychology, 42(1). https://scielo.org.za/scielo.php?script=sci_arttext&pid=S2071-07632016000100010
Keep reading
Assessment

"Airtime is not contribution in a group exercise"

"The candidate who talks most usually wins the group exercise, and that is a measurement error rather than a judgement. A contribution map with four countable moves separates chairing the discussion from filling it."

Assessment

"An in-tray exercise is not a speed test"

"Most in-tray scoring rewards the number of items a candidate touches, which makes a prioritisation test into a typing race. A three-way triage rule and a delegation standard fix the answer key."

Want this applied inside your organisation?

Talk to us