A score is not a hiring decision

The summary is not the substance

A single overall score is tempting. It sorts a list, it fits in a spreadsheet and it feels objective. The trouble is what it compresses. Two candidates can arrive at the same number by entirely different routes.

Imagine two applicants with identical overall scores. One has deep technical knowledge and struggled to explain it simply. The other communicated beautifully and was vague on the detail. For a role that faces customers, the second may be the stronger hire. For a role deep inside a system, the first. The number cannot tell you which, because it does not know what you need.

A score is a pointer. It tells you where to look. It is not the thing you are looking for.

Numbers also carry a false precision. A gap of a few points between two candidates can look decisive on a sorted list while meaning almost nothing about the people.

Put the words beside the number

The most useful change to a candidate report is to attach evidence to every judgement. Not a paragraph of generated praise. The candidate's own words, quoted, under the score they support.

Suppose a report rates a candidate's reasoning as solid. Beneath it sits the sentence they actually said about how they diagnosed a slow process. A reviewer can read that line in seconds and decide whether they agree. If they do not, they can say why. That is a review. Accepting the number without the line is not.

Evidence also makes a report fairer to the candidate. A judgement that can be checked can be challenged. A judgement that cannot be checked simply has to be trusted.

A useful habit is to read the quote first and guess the score before looking at it. Where your guess and the report disagree, you have found something worth a second look.

A score you can disagree with in seconds is a score you can stand behind.Risenoow Editorial

Dimensions tell you more than totals

Scoring several distinct dimensions separately keeps useful detail in view. Knowledge of the role, quality of reasoning, clarity of communication, whether answers held up under follow up and genuine interest in the work are different things. Folding them into one figure loses the shape of the person.

Dimensions also show where to probe next. A candidate strong on knowledge but uneven when pushed for specifics is a clear brief for the second interview. Ask who made the decision they described. Ask what they would have done if it had failed.

Each dimension should carry a band as well as a figure, plus a plain sentence explaining it. Reviewers read sentences more carefully than numbers.

Some things should never be scored at all. Accent, grammar, vocabulary and filler words are not evidence of capability. Personality and emotion should never be inferred. A report that scores any of these is measuring the wrong thing.

Set the interview against the paperwork

A CV and an interview are two readings of the same person. They are most useful when they meet. Where the interview supports a claim on the CV, confidence rises. Where the two disagree, you have found the most valuable question for the next conversation.

Imagine a written answer that claims the candidate led a migration. In the interview, they describe it as a project they joined halfway through. Neither statement is a lie on its own. Together they are a clear prompt: ask what they personally decided.

Mark each claim as supported, unclear or contradicted. An unclear claim is not a mark against the candidate. It is a note that the evidence has not yet settled the question.

This cross check is one of the most valuable things a report can do, because neither source on its own can reveal a contradiction.

• Supported: the interview confirms what the documents state.
• Unclear: the conversation did not establish it either way.
• Contradicted: the two sources disagree and the next interview should ask why.

Know when not to trust the number

Some applications contain very little. A short interview, a sparse CV, a handful of written words. A report that still produces a confident score is inventing precision it does not have.

The honest approach is to withhold the score and say why. State how much material there was. Make clear that what follows are notes, not a measure. A thin application is not a thin candidate. It is an unfinished picture.

Comparisons deserve the same care. Placing a candidate against the average of other applicants can help, but only when enough people have been measured for the average to mean something. A practical reference point is the lowest score the team has already shortlisted for the role: the bar set in practice rather than in theory.

Withholding a score can feel unhelpful in the moment. It is far more helpful than a confident figure that later has to be explained away.

Decide from the evidence, then write it down

Once the evidence has been read, the decision belongs to a person. Write the reason in terms of the work, not the number. 'Strong on the systems this team runs, but could not explain who owned the target' is a reason. 'Scored lower' is not.

A written reason also improves the process over time. Reading back through a season of decisions shows which dimensions mattered in the role and which the team overvalued.

A Risenoow report scores five dimensions and quotes the candidate's words under each one, checks claims against the CV and gives no score when there is too little evidence to judge. The decision itself stays with a person on the hiring team.