Resume Screening Criteria: How to Write Them

Short answer: Resume screening criteria are the written, job-related standards every applicant is measured against, sorted into knockouts (hard requirements the role cannot work without), must-haves (scored requirements), and preferred signals (tie-breakers). Write them before the posting goes live, keep every criterion tied to something the job actually needs, and apply the same list to every resume in the pile.

Last updated July 2026.

What are resume screening criteria?

Resume screening criteria are the specific, job-related standards a recruiter checks each application against before deciding who moves forward. Good criteria are written down before applications arrive, phrased so two different reviewers would reach the same answer, and tied to a requirement of the role rather than a preference about the person.

Most screening problems are criteria problems wearing a disguise. When a hiring manager says the pile is bad, what usually happened is that nobody agreed on what "qualified" meant until after forty people applied, so the first fifteen resumes were judged against one standard and the rest against another. Writing the criteria first is unglamorous and it fixes more than any tool will.

The four tiers of screening criteria

Almost every workable screening rubric sorts into four tiers. The distinction that matters most is between a knockout and a must-have, because teams routinely promote a preference into a knockout and then wonder why the shortlist is empty.

The tier a criterion belongs in is a business decision, not a technical one. A hospital screening for an unlicensed nurse has no discretion. A software team requiring a computer science degree usually does, and treating that preference as a knockout is how good candidates get filtered out by a rule nobody would defend out loud.

How to write a knockout criterion that holds up

A knockout criterion should pass three tests. First, it has to be job-related: you can point at the work and explain why the job cannot be done without it. Second, it has to be observable in an application, not inferred. Third, it has to be phrased so two reviewers reading the same resume give the same answer.

"Strong communication skills" fails all three. "Has held a role requiring written client updates" passes the first two and mostly passes the third. "Five plus years experience" passes the third test and often fails the first, since the number is usually a shorthand for a scope requirement nobody wrote down. If the real requirement is that someone has closed a month-end alone, write that instead. You will get a better pile and a defensible record of why.

Compensation expectation deserves its own note, because it is one of the most common screening criteria and one of the easiest to apply badly. Screening someone out for asking above the range only works if you have a defensible pay band for the role in the first place, and in the growing number of US states with pay transparency rules you are publishing that range anyway. Screen against the posted band, not against whoever asked for the least.

A resume screening checklist you can copy

This is the working order most recruiting teams converge on. It is deliberately front-loaded: the cheap, binary checks run first so human attention lands on the applications that survive them. If you want the scoring mechanics rather than the criteria themselves, the companion piece on running a resume screening checklist covers the three-point scale and how to keep two reviewers landing in the same place.

1. Confirm the eligibility basics. Work authorization, licensure where the role requires it, location or onsite requirement, and any legal minimum such as age for interstate driving.
2. Run the knockouts. No more than three or four. Flag misses with a reason rather than deleting the application, so you can revisit if the pile comes in thin.
3. Check scope, not just years. Read for the size of the thing they ran: team size, budget, transaction volume, patient load, territory. Years of experience is a weak proxy for all of them.
4. Score the must-haves against the same rubric. Same weights for every applicant, written before the pile arrived.
5. Look for corroboration. Named systems, specific numbers, and a coherent progression. Vague resumes are sometimes weak candidates and sometimes just badly written ones, so treat this as a light signal.
6. Note the gaps you would ask about, rather than deciding what they mean. Employment gaps have many explanations and assuming the worst one is both unfair and frequently wrong.
7. Rank, then cut. Order the pile by score, then draw the line based on how many people you can actually interview this week.
8. Record why. One line per candidate on the criteria they met and missed. This is what makes the process reviewable six months later.

Steps one and two are where automated screening earns its keep, because they are mechanical, repetitive, and exactly the part humans do worst when they are tired. Steps three through six are judgment, and they are the reason candidate screening software should rank and flag rather than reject on its own.

Resume screening methods compared

The honest summary is that the method matters less than whether criteria exist. A structured rubric applied by a careful human beats a badly configured tool. What a tool changes is whether the rubric still gets applied to resume number 300 as carefully as to resume number 3, which is where automated resume screening pulls ahead at volume.

What you should not screen on

Screening criteria live under employment law, and the statutes are unchanged: Title VII, the ADA, and the ADEA all still apply, and so does disparate impact liability. A criterion does not have to be intentionally discriminatory to create a problem. It has to disproportionately screen out a protected group without being job-related and consistent with business necessity.

The practical version for a recruiter: avoid criteria that function as proxies. School prestige, continuous employment history, ZIP code, graduation year, and photo or name based judgments all correlate with things you are not allowed to hire on and rarely predict performance. Keep criteria tied to the work. If you use an automated tool, several states and cities now regulate automated employment decision tools directly, New York City's Local Law 144 being the most cited, so check what applies where you hire and keep a human making the decision.

How specific should criteria get for a niche role?

More specific than feels comfortable, as long as every item is a real requirement. Generic criteria produce generic shortlists. A carrier screening CDL applicants does not check "driving experience," it checks class, endorsement, months of verifiable over the road time, and home time preference, which is why truck driver recruiting software screens on endorsements rather than a keyword. The same logic applies to a licensure-driven clinical role or a district hiring by certification area: the criteria that actually sort a pile are the ones written in the vocabulary of that job.

How long should screening one resume take?

A trained recruiter working from written criteria typically spends 30 seconds to 2 minutes on a first-pass screen, and longer on the applications that survive it. The number matters less than what it implies at volume: a 200-application posting is somewhere between two and seven hours of concentrated reading, which is why the realistic ceiling on resumes a recruiter can screen is the constraint most teams are actually fighting.

What is the difference between screening and evaluation?

Screening decides who gets a closer look. Evaluation decides who gets hired. Screening is fast, criteria-based, and applied to everyone; evaluation is deep, involves interviews and work samples, and is applied to a shortlist. Conflating them is a common and expensive mistake, because it pushes teams to make hiring-grade judgments from a two-page document that was never designed to support them.

Can resume screening be consistent across reviewers?

Only if the criteria are written down and weighted before anyone reads a resume. The single highest-return change most teams can make is to hold a 20 minute meeting with the hiring manager before the posting goes live, agree the knockouts and the top three must-haves, and write them in one place. Everything else, including whether you automate, is downstream of that. If you want the mechanics of turning those criteria into questions the application itself asks, eligibility and knockout screening is the layer that does it.

Where to start

Pick your next open role and write four things: the knockouts, the top three must-haves with weights, the preferred items you are explicitly not filtering on, and one line describing what "good" looks like in the first 90 days. That document is your screening criteria. Everything after it, manual or automated, is just applying it consistently to every person who applied.