Why \\"ATS Optimisation\\" Tools Sell You a Problem

You upload your CV, wait a moment, and receive a score. 62 out of 100. Below the threshold. Here is what to fix, and here is the subscription. A broader overview of employer-side monitoring tools is available through employee monitoring software.

The score is not measured against anything real, the threshold is invented, and the filter it claims to model mostly is not running. For an independent public reference, consult NIST AI Risk Management Framework.

We build a resume product. That makes this an awkward article to write and it is the reason it is worth writing — the honest version of what these tools can do is narrower than the category claims, including for ours.

What the score is

It is not produced by an applicant tracking system. The tool has no access to the employer's system, does not know its configuration, and cannot know what the recruiter searches for.

It is the tool's own model of what such a system might do — usually keyword overlap against the job description, plus formatting rules, plus length and section checks.

The threshold is chosen by the vendor. There is no industry pass mark. A score of 62 means the vendor's formula produced 62.

And the formula is designed to produce a failing score initially, because a tool that congratulates you has nothing to sell.

Why the underlying premise is wrong

The category rests on the belief that these systems automatically reject most applications on content.

Recruiters interviewed in a 2025 study reported that only 8% had configured content-based automatic rejection, while 92% rejected manually or through knockout questions alone.

The "75% of resumes rejected" figure traces to a 2012 sales pitch from a company that has since gone bankrupt.

And around a fifth of recruiters, asked where the myth came from, pointed at resume-writing services and career coaches marketing "ATS-optimised templates" through fear-based messaging.

The tools are a named source of the belief they are sold to address.

See what applicant tracking systems actually do.

What the tools get right

Being fair, because some of the advice is sound even where the framing is not.

Keyword overlap is a real signal, for a real reason: recruiters search the parsed database by keyword. A CV that never uses the words your field uses is harder to find. That is a searchability point, not a rejection point, and the difference matters — it means you want the right words present, not the maximum number of them.

Simple structure parses better. True, and worth doing.

Missing sections and unclear dates genuinely make a CV harder to read, by a person as much as by a parser.

A second look at your CV is useful. The problem is what the second look is framed as.

Where they cause harm

Wasted effort. Time spent raising a meaningless score is time not spent on the things that do determine outcomes — applying early, answering screening questions correctly, targeting better, using referrals.

Worse CVs. Optimising for keyword density produces documents that read badly to the person who eventually opens them. A CV that scores 95 and bores a recruiter has failed at the only test that runs.

Dangerous advice. Some tools have suggested invisible keywords, white text or hidden blocks. Recruiters find these, and the reaction is not neutral — it reads as an attempt to deceive, because it is one.

Misattributed failure. The most damaging effect. A candidate who believes the software rejected them does not examine whether they applied late, missed a knockout question, or were the four hundredth applicant. The myth prevents diagnosis.

And it is demoralising in a specific way. Being rejected by a person for a reason is difficult. Believing you were discarded unread by a machine, repeatedly, is worse and it is mostly not true.

What a resume tool can honestly do

The narrower and genuinely useful list.

Check that it parses. Show what a parser extracts from your file. If your job titles land in the wrong field, that is a real, checkable problem.

Check completeness and clarity. Missing dates, unexplained gaps, vague descriptions, achievements with no outcome.

Compare against the job description as a prompt for thinking, not as a score. "This role emphasises X and your CV does not mention it — do you have that experience?" is useful. A percentage is not.

Help you write the difficult parts. Quantifying an achievement, describing a career change, explaining a gap. See quantifying achievements.

Tailor without rewriting from scratch, which is the actual time cost of applying properly.

Practise interviews, which is where preparation is most under-done and where feedback is hardest to get.

None of that requires a fictional score, and a tool that offers a score is telling you something about its priorities.

Evaluating any tool in this category

Does it give a score against an unnamed standard? Ask what the standard is. There is no answer.

Does it claim to know what "the ATS" wants? There is no single ATS, and it does not have access to any of them.

Does it suggest anything hidden from the reader? Stop.

Does the free version exist to produce a failing result? Look at whether anyone ever passes.

Does it cite sources for its claims about hiring? Almost none do, and the claims are almost all the same three unsourced statistics.

And what does it do with your CV? You are uploading your employment history, contact details and dates. Where does it go, how long is it kept, and who else receives it. See our own privacy policy — we send content to a model provider when you ask for text to be generated, and we say so.

The short version

The score is the vendor's formula against an invented threshold, for a filter that recruiters report mostly is not running.

Keyword overlap matters for searchability, not for rejection — you want the right words, not the most.

The real harm is misattribution. A candidate who blames the software stops looking at timing, screening questions and targeting.

A resume tool can honestly check parsing, completeness and clarity, help with the hard sentences, and prepare you for interviews. It cannot tell you what a system it has never seen will do.

And check what any of them does with your CV, including ours.