Your ATS score is not a pass or fail grade

A candidate uploads the same CV to three online ATS checkers. One gives it 42%, another gives it 76%, and the third calls it an excellent match. Which score is correct? Possibly none of them. The uncomfortable truth is that there is no universal ATS exam and no single score that every employer uses.

An applicant tracking system is not one machine with one set of rules. Employers configure different platforms, workflows, questions, filters, and hiring stages. A score from a CV tool can still be useful, but it is best treated as a diagnostic signal: it can point to missing language, weak evidence, or parsing risk. It cannot promise that an employer's system will rank your application in exactly the same way.

The useful question is not, "Will my CV beat the ATS?" It is, "Can the system read my CV, can it find evidence relevant to this job, and will a recruiter understand that evidence quickly?" Those are three separate tests. A strong application needs to pass all three.

Myth 1: Every ATS automatically rejects low-scoring CVs

Some hiring workflows use ranking, search, matching, or screening rules. Others mainly store applications and help recruiters organize them. A candidate may also be excluded by a knockout question before the CV becomes important: work authorization, location, a required licence, salary expectations, or willingness to work a particular schedule.

This matters because candidates often blame formatting or keywords for a rejection that may have had a different cause. No CV rewrite can solve an eligibility mismatch. Before polishing individual sentences, check the non-negotiable requirements and answer application questions carefully.

Myth 2: More keywords always produce a better result

A keyword is valuable when it helps identify genuine experience. Repeating the job title fifteen times or hiding a block of white text does not make a candidate more qualified. It can make the CV unpleasant to read, and suspicious language may weaken trust when a person opens it.

Think in terms of evidence coverage instead. If a vacancy repeatedly asks for stakeholder management, do not merely add the phrase to Skills. Show where you used it: who the stakeholders were, what you needed from them, what you communicated, and what happened as a result. One credible bullet can carry more weight with a recruiter than five disconnected keyword mentions.

Weak: "Stakeholder management, reporting, communication, project delivery."

Stronger: "Coordinated weekly delivery reviews with product, finance, and customer teams, resolving ownership gaps and bringing three delayed implementation milestones back on schedule."

The stronger version contains relevant language, but it also gives that language a setting. It tells a human reader what the candidate actually did.

Myth 3: An ATS-friendly CV must be plain and lifeless

Readable design and machine-readable structure are not enemies. A CV can have hierarchy, considered spacing, restrained colour, and a clear visual identity while keeping its core text easy to extract. Problems arise when important information depends on decorative icons, images, floating text boxes, unusual reading order, or labels the parser cannot interpret.

The safest approach is conservative where meaning is involved. Use recognizable section headings. Keep employer, title, and dates clearly associated. Use real text for contact details. Avoid placing essential qualifications only inside graphics. A polished layout should make the structure clearer, not ask the reader or parser to solve a puzzle.

The three tests that matter more than a perfect score

Test 1: the extraction test. Open the exported PDF, select all the text, and paste it into a plain-text editor. Is the name first? Do headings, job titles, employers, dates, and bullets appear in a sensible order? Are any lines missing or duplicated? This does not reproduce every ATS parser, but it reveals many avoidable export problems.

Test 2: the evidence test. Take the five most important requirements from the vacancy and find proof for each one in the CV. Not just the word: proof. If the advertisement asks for forecasting, the reader should find what you forecast, which tools or data you used, and ideally the scope or outcome. Where you have honest evidence but use different terminology, align your wording with the employer's language.

Test 3: the ten-second test. Show the top half of the first page to someone who has not seen the vacancy. Ask what role the person is targeting and what their strongest relevant qualifications are. If the answer is unclear, the problem is probably emphasis rather than ATS compatibility. Improve the summary, ordering, headings, and first few bullets before adding more keywords.

Why two ATS checkers disagree

Different tools choose different keyword lists, weight repeated terms differently, interpret seniority in different ways, and make different assumptions about formatting. Some compare vocabulary. Some inspect section structure. Some reward measurable achievements. Some penalize missing contact or education fields. Their percentages may look precise even though the underlying judgments are not identical.

Use a score comparatively, not absolutely. Run the same CV against a specific vacancy, improve weak but truthful areas, and check whether the result becomes more relevant and readable. Do not distort a strong CV merely to move an unexplained number from 84 to 97.

A better pre-application checklist

Before applying, confirm that the exported file opens correctly, its text can be selected, and no content is clipped. Use standard headings and consistent dates. Match the job's important terminology where it accurately describes your experience. Put the strongest matching evidence on the first page. Remove unsupported skills and repeated claims. Then read the document as a recruiter would: quickly, skeptically, and in search of proof.

The goal is not to trick a gatekeeper. It is to reduce the distance between what the employer needs and what your CV proves. When parsing, relevance, and human readability all support the same story, the exact ATS score becomes far less important.