A repeatable resume screening checklist cuts CV review time roughly in half, standardises the way every panel member evaluates a candidate, and gives you a defensible audit trail when a rejection is questioned. This 15-point checklist covers the pre-screen setup, the 30-second first pass, the deep review of survivors, the 5 red flags to watch for, a bias sanity check, and a GDPR-safe rejection process. Print this page (Ctrl/Cmd + P → Save as PDF) and hand it to your hiring panel - the whole thing fits on one page.

Below is the checklist plus enough context to explain each step. If you already know what you're doing, jump straight to the numbered list.

Why a checklist, and why now

Recruiters spend an average of four hours per day on CV review. Most of that time is wasted on candidates who should have been rejected in the first thirty seconds. A standardised checklist prevents four things: order bias (the top of the pile getting more attention than the bottom), fatigue variation (a tired reviewer scoring differently than a fresh one), inconsistent criteria across panel members, and the vague-verdict problem when a rejection is challenged. It also cuts the raw time cost by roughly half, because the deep review only happens on candidates who survived the fast first pass.

Before you open a single CV - set the criteria

The most important step in the whole process happens before you touch a CV. Write down three things. Nothing more.

1. Pick the three must-haves from the job description. One line each. If you have eight "must-haves," you don't have a role, you have a wishlist. Cut it down.

2. Assign a weight to each. Something like 40/30/30 or 50/30/20. This becomes your scoring rubric.

3. Decide the reject threshold in advance. For example: anything scoring below 60 out of 100 is out. Fixing the threshold before you look at any CV protects you from talking yourself into a marginal hire because you liked their university.

4. Nominate the human reviewer for every reject decision. Under GDPR Article 22 and the equivalent UK provisions, a candidate has the right not to be subject to a decision made "solely by automated means" when it "significantly affects" them - and rejecting a job application does. Assign a human to look at every reject before it goes out. See our GDPR guide for AI CV screening for the full compliance picture.

The 30-second first pass

Skim, don't read. Aim to cut the pile by 60–70% at this stage.

5. Format and readability check. Clean layout, real headers, no walls of text. If the CV is unreadable to you, it's unreadable to your team.

6. Must-have keyword scan. Are the three must-haves you defined earlier actually present? Not "similar to." Present.

7. Recency check. Is the most recent role in the right seniority band? A senior title from ten years ago is not the same as a senior title today.

8. Location and right-to-work. Legally required to hire? Willing to relocate or work in the office pattern you need?

9. Employment continuity. Gaps longer than six months are worth flagging (not rejecting). Ask about them at interview.

Timebox this: 30 seconds per CV. Fifty CVs takes twenty-five minutes.

The deep review - 5 to 8 minutes per candidate who passed

Now spend real time on the survivors. Typically fifteen to twenty CVs make it here.

10. Skills relevance. The must-have skills are not just present - they are recent, they are demonstrated in context, and they are used at roughly the seniority the role requires.

11. Impact evidence. Numbers, outcomes, ownership. "Increased conversion by 22%" beats "worked on conversion optimisation." Look for the specific.

12. Career trajectory. Is the trajectory upward, lateral, or unclear? Lateral is fine at senior levels. Unclear needs a question at interview.

13. Domain fit. Same-industry moves are lower-risk than cross-industry moves. Cross-industry is not disqualifying - many great candidates come from adjacent domains - but note it and ask about the transition rationale.

14. Signals of ownership. "Led a project" vs "contributed to a project" is a real signal. Look for words like led, owned, launched, decided, hired, shipped.

Score each candidate on the rubric you defined at step 2. Write two lines of recruiter-ready notes per candidate - highlights and questions to ask.

The 5 red flags to watch for

Not automatic rejections. Flags to investigate.

Job-hopping - more than three roles in less than three years without clear progression.
Skill inflation - senior title, junior responsibilities described.
Copy-paste generic CV - no tailoring to the role, no mention of specifics the JD asked for.
Chronology inconsistencies - dates that don't add up, employers that overlap without explanation.
Unexplained "career break" longer than 12 months - often completely fine (family, health, sabbatical), but ask.

The bias sanity check - step 15

15. Anonymise the first pass, audit the shortlist, escalate imbalance. Redact name, photo, date of birth, and address at the first-pass stage. Once a quarter, compare the demographic distribution of your shortlist to your applicant pool. If ninety percent of your top ten look like your existing team, question the job description before you question the candidates. See is AI CV screening biased? for the deeper mechanics.

How to reject fairly

Every reject decision needs a human reviewer (Article 22). Log the reason in one sentence. Send a proper rejection email today, not "when you have time" - silence is the worst answer and it destroys employer brand. A three-line, warm, generic rejection is better than a bespoke essay that never gets written.

Retain the CV plus the reject reason for six to twelve months. That covers the discrimination-claim window in most jurisdictions.

The AI shortcut

Everything above can be run manually with this checklist, or automated. Manual on fifty CVs takes about three hours end-to-end (twenty-five minutes first-pass plus two hours deep review plus rejection emails). With an AI CV screening tool, the same fifty CVs are scored on the same fifteen dimensions in about two minutes, and you spend fifteen minutes reviewing the ranked shortlist plus writing rejection notes. Same fifteen dimensions. Same defensible audit trail. Different unit-time cost.

If you want to test the AI path on one CV to see what the output looks like, try SwiftShortlist free - no signup for the first CV.

The checklist - print this section

Copy or print the numbered list below and hand it to your hiring panel.

Pick the three must-haves from the JD. One line each.
Assign a weight to each must-have (e.g. 40/30/30).
Set the reject threshold before opening any CV (e.g. score below 60).
Nominate the human reviewer for every reject (GDPR Article 22).
First-pass: format and readability check.
First-pass: are the three must-haves actually present?
First-pass: most recent role in the right seniority band.
First-pass: location and right to work.
First-pass: flag gaps over six months (don't reject).
Deep review: skills relevant, recent, demonstrated in context.
Deep review: impact evidence - numbers, outcomes, ownership.
Deep review: career trajectory - upward, lateral, or unclear?
Deep review: domain fit - same industry, adjacent, or unrelated?
Deep review: signals of ownership - led, owned, launched, decided.
Bias sanity check - anonymise the first pass, audit the shortlist quarterly.

Print this page (Ctrl/Cmd + P → Save as PDF) for a one-page reference. See also the cost-per-hire calculator to model what the checklist saves you in recruiter time, or read the complete guide to AI CV screening for the wider picture.

The Resume Screening Checklist Every Recruiter Should Use in 2026