What is a hiring scorecard and how do I make one?

Most job descriptions describe activity: manage the team, oversee operations, liaise with vendors. A scorecard describes results: what will be true in twelve months if this hire worked. That single change is what makes it useful, because you can score a person against a result and you cannot score them against a verb.

It fits on one page. At the top, one sentence on why this role exists. Under that, three to five outcomes with a number and a date where you can put one. Under those, four to six criteria, each traceable to an outcome, each with a 1-3 scale and a written description of what a 1 and a 3 look like. That is the whole document.

The scorecard has to exist before the first applicant arrives. Written afterwards, it becomes a justification of the person you already liked, which is worse than having none at all because it wears the costume of rigour. Written first, it is the thing every stage evaluates against, and it is the reason two people on your team can screen the same pile and land in the same place.

It is also the cheapest disagreement you will ever have. Two founders who cannot agree on the four outcomes for a new operations hire were never going to agree on the candidate. Better to find that out over a blank page than over a person you have already made an offer to.

Before you write anything about a person, answer why the role exists at all. Not the polite version. The specific one: deliveries are late and the founder is personally chasing drivers, or the monthly close is eating the first two weeks of every month, or nobody owns whether customers renew.

Then write what must be true in a year for the hire to have worked. Three to five outcomes, each a result rather than an effort. Put a number and a date on the ones that carry one, and accept a qualitative outcome where a number would be fake. An outcome that cannot be checked in twelve months is a wish, not an outcome, and it belongs in the bin.

The test of a good outcome is that a reasonable person could say at the end of the year whether it happened. Improve operational efficiency fails that test. Cut late deliveries from one in five to below one in twenty passes it.

Write outcomes for the role you have, not the role you imagine. A five-person company hiring its sixth does not need a strategy outcome; it needs the thing that is currently on fire to stop being on fire. Overreaching here is how you end up interviewing people who are wrong for the job you actually have.

Now go outcome by outcome and ask what a person has to be able to do to produce it. Those abilities are your criteria. If a criterion does not trace back to a named outcome, remove it. This is how you keep the list at four to six instead of drifting to fifteen, and it is how you avoid the classic trap of scoring people on things that sound impressive but do not move any result you care about.

Three criteria are worth assessing on every role, at very different bars. Whether the person sees the situation clearly. Whether they act on it without being pushed. Whether they carry other people along. In plainer language: systems thinking, agency, and communication.

These are universal in kind and variable in degree. A dispatch rider and a finance director both need systems thinking; the bar is nowhere near the same. So the role sets the bar and the score is read against that bar, never floating on its own. A 2 on communication for a back-office analyst and a 2 for a customer-facing manager are two different verdicts.

Everything else on the scorecard is role-specific and comes from your outcomes. Keep the language consistent across roles you hire for, so that a person you assessed for one role is still comparable when a different role opens. Two different names for the same quality is how a scorecard quietly stops being comparable.

• Systems thinking: sees how the parts connect, spots the real constraint rather than the loudest symptom.
• Agency: moves without being told, decides under uncertainty, owns the outcome rather than the task.
• Communication: makes other people understand and act, upward and downward, in writing and in the room.
• Role criteria: drawn from your outcomes, four to six total including the constants, each one observable.

Score on 1-3, or 1-4 if you genuinely need a distinction between adequate and strong. Anything wider is false precision. Nobody can tell a 6 from a 7 on a hundred-point rubric, and the scores end up clustering in the middle anyway while taking three times as long to produce.

Anchors are the part that makes the scale mean anything. For each criterion, write in one sentence what a 1 looks like and what a 3 looks like, in terms of observable evidence rather than adjectives. Not strong communicator. Something like: explained a decision to a non-technical stakeholder in writing and got a decision out of it.

Score evidence, not impressions, and write the evidence next to the score. One line is enough: the claim, the source, why it earned that number. A score with no evidence beside it is a feeling with a number on it, and it will not survive the conversation where two people disagree.

Do not average the scores into a single verdict and stop thinking. An uneven profile is often the most informative thing on the page. A candidate at 3, 3, 1 has a specific gap you can decide about; a flat 2 across the board usually means you learned nothing about them.

1. Write one sentence on why the role exists.
2. Write three to five outcomes, each checkable in twelve months, with numbers and dates where they are real.
3. For each outcome, name the ability that produces it. Those are your role criteria.
4. Add the three constants and set the bar each one needs for this specific role.
5. Cut the list to four to six criteria. If it will not cut, your outcomes are too many.
6. Write 1 and 3 anchors for every criterion, in observable terms.
7. Decide which stage gives you evidence for which criterion, so no criterion is left unassessed.
8. Score the first twenty candidates with a second person independently, then compare and fix whichever criterion you disagreed on.

The company runs same-day deliveries across Lagos with twenty riders. The founder is personally chasing failed deliveries every evening, drivers are managed over WhatsApp, and the two biggest clients have started asking about reliability. That is why the role exists.

Outcomes: cut failed same-day deliveries from roughly one in five to below one in twenty within nine months; move dispatch off ad hoc WhatsApp onto one system the whole team uses daily by month four; keep rider attrition under a defined threshold through the year; make the founder unnecessary to daily operations by month six.

Criteria derived from those: route and dispatch design under real Lagos constraints, which traces to the failure-rate outcome. Rider management and retention, which traces to attrition. Systems thinking, because moving twenty riders off WhatsApp onto a working process is a systems problem before it is a software one. Agency, because the founder stepping back only works if the person moves without being asked. Communication, because the whole job is getting riders and clients to do things.

That is five criteria, each with an owner outcome. A 3 on rider management reads: has directly managed a team of fifteen or more field staff, can describe what they changed to reduce attrition, and what it cost. A 1 reads: has coordinated field staff only through someone else. You now have something you can score a hundred applicants against, and something you can defend to a co-founder who preferred a different candidate.

The most common failure is the hundred-point rubric with fourteen weighted sub-criteria. It looks rigorous and it is unusable. Nobody scores it consistently, the weights are guesses, and the total is a number with no meaning. Six criteria that everyone actually applies beat fourteen that nobody does.

The second failure is criteria nobody can observe. Cultural fit, passion, drive, professionalism. These are not assessable; they are the vocabulary people reach for when they want to justify a preference after the fact, and they are where bias enters a process that looks structured. If you cannot name the evidence that would move the score, the criterion is not real.

The third is scoring the CV instead of the evidence. The CV is one claim among several, written by the person being assessed, and optimised for exactly this moment. Score what they wrote in the application, what their history holds up under questioning, and what they did when they could not prepare.

Tezera can draft a scorecard for you: you say why you are hiring and what must be true in a year, and it proposes the outcomes, criteria and anchors for you to trim, then scores every candidate against them with the evidence quoted. The limitation is the honest one. It cannot skip the first step. If you have not defined what the role must deliver, there is nothing to score against, and no tool will invent your outcomes for you.

The same question, phrased the other ways people ask it. Every one of them lands on this page.

• How do I create a hiring scorecard for a role?
• What should a candidate evaluation scorecard include?
• What is a scorecard in recruitment?
• How do I score candidates objectively?

• →How do I screen hundreds of job applications quickly?
• →Why do new hires quit in the first three months?
• →Operations manager job description generator
• →Operations manager interview questions (Nigeria)

Everything else we have written like this sits on the answers index.