What applicant tracking systems actually do with your resume

There is a lot of advice built on the idea that software grades your resume and throws most applications away before a person sees them. That is not how these systems work, and believing it leads people to write worse resumes. Here is the accurate version.

What the software does

An applicant tracking system is a database. It takes the document you upload, converts it into structured fields such as your name, your employers, your job titles and your dates, and stores the result so a recruiter can search it later. When a role gets four hundred applications, that search is how anyone finds anything.

Some of these products also let a recruiter sort or rank results by how well a record matches a set of keywords. That feature is real. What it produces is an ordering of candidates for a person to work through, not a verdict.

What the software does not do

It does not assign your resume a grade. There is no quality mark attached to your document, no passing threshold, and no hidden number you can raise by loading the page with the right words.

The number everyone quotes for this does not exist. The claim that 75 percent of resumes are rejected by software before a human sees them traces to a single vendor called Preptel, which put it in a sales pitch in 2012 and never published a study, a data set or a method. The company shut down in 2013. The figure survived it by being repeated, from Forbes in 2014 to CIO in 2018 to CNBC in 2019, each citing the one before rather than the origin.

What it does do, which is not the same thing

Employers configure filters, and those filters do remove people before any human reads an application. This is the part the myth gets directionally right and mechanically wrong.

Harvard Business School studied this in 2021, in a report called Hidden Workers: Untapped Talent. Joseph Fuller and his co-authors found that around 99 percent of Fortune 500 companies run this software, and that its filters exclude an estimated 27 million people in the United States who could do the work. One finding in particular is worth knowing if you have ever taken time away from work: half of US companies filter out applicants with a gap in their work history longer than six months. Degree requirements and years of experience are used the same way.

So the exclusion is real, but it is a rule an employer set, applied to the fields your document was converted into. It is not a judgement about how good your resume is. Knockout questions on the application form itself, about work authorisation or a required certification, work the same way.

The practical consequence is narrower than the myth and more useful. You cannot write your way past a filter that was set before you applied. What you can control is whether your dates, employers and titles convert into that record correctly in the first place, because a filter reading a mangled record is reading the wrong thing about you.

The second filter, and the one you can still win

Clear the employer's rules and you reach the part that decides most outcomes. A person opens the search results and moves through them quickly, looking for whether your recent work resembles the role they are filling.

How quickly is worth being careful about, because the usual numbers are shakier than they look. The six or seven seconds everyone quotes comes from a 2018 eye-tracking exercise run on thirty recruiters by a company that sells resume services, and its method was never published. A larger 2024 survey of 418 hiring professionals put the first pass nearer eleven seconds, with most reviewers spending between ten seconds and a minute. The honest summary is that it is a skim rather than a reading, and that is the only part of it your resume has to survive.

That first pass sorts rather than decides. It is not the interview decision, which comes later and usually involves someone else. It decides whether you stay in consideration, and it runs in one direction only: passing it buys you a longer read where your detail gets weighed, while failing it ends the application there. Almost every application is lost at this stage, and it is the only stage your writing reaches at all. A filter set before you applied cannot be argued with. Whether the strongest thing about your experience is visible immediately, or sitting further down under a list of responsibilities, is entirely yours.

Both problems are structural, and structure is fixable. A clean document that converts properly is the baseline. Getting your best material to the top is what changes the outcome.

That is what the builder is for. It is one format, refined over fifteen years of IT delivery work, and it puts a Professional Summary of your experience and highlights directly under your name at the top of the first page, as three to five quantified bullets. Experience follows in reverse chronological order, then education. Single column, no tables, no text boxes, so the structure survives conversion into a searchable record. The order is the point: the reader does not have to go looking for your strongest work.

Why this page exists

This product used to describe its own metric in the language everyone else uses. In July 2026 several recruiting professionals pointed out on LinkedIn that the framing was wrong, and they were right. The wording was corrected across the product and the metric is now called the Resume Readiness Score, which is what it always measured: how cleanly your resume converts, and whether your strongest content leads.

Common questions