What happens to your resume between the submit button and the recruiter's screen
Every job-search thread lands on the same explanation: a robot binned your resume before a person ever saw it. I build careerify, which gives the ATS score away free and charges for the tailored resume afterwards, so keeping that fear warm would suit me commercially. The mechanism is duller and far more useful: an applicant tracking system takes your application in, converts the document into text and fields, lets a recruiter search the pile, and ranks it. Nowhere in that chain is a lever marked reject that reads your layout and forms an opinion about you. This page walks through what happens to the file, what really screens you out, and which part of that a score can help with.
What the system does with the file you submit
An applicant tracking system is unglamorous software for handling a pile of applications. It takes them in from the careers page and the job boards, converts your document into text, and tries to map that text into fields such as name, employer, title, dates and skills. Then it lets a recruiter search and filter the pile, sort it, move people between stages and send templated emails when something changes status. That is a database with a workflow on top, and those four verbs are the whole mechanism: take in, parse, search, rank. What any particular vendor does inside it I cannot tell you, because setups differ by company and by role and none of us outside those companies are reading their logs. What I can tell you is what happens when your layout fights the parse, two columns, text that is really an image, job titles sitting in a header: content lands in the wrong field or never makes it into the text at all. The result is not a verdict, it is an absence, you do not come up in a search you should have come up in.
Hidden text is not hidden from the machine
The first thing that happens to your file is extraction: it is reduced to characters in order, while colour, font size and layer order are instructions for drawing those characters rather than part of them. That inverts the shortcut people reach for once they believe in the auto-reject, pasting the ad in white one-point type or hiding an instruction addressed to a language model. To the parser the block is ordinary text, which makes it ordinary keyword stuffing, and it is invisible to you when you proofread while staying perfectly visible to any recruiter who copies your resume into a plain text field. I build one of these readers, so I had to take a position: careerify's free check puts your CV and the job ad inside named blocks with a system rule stating that everything inside is data and never an instruction, and it defangs the block tags so a CV cannot close its own block and speak from outside the data. The comment sitting directly above that code in our repository calls it a mitigation rather than a guarantee, and that is the honest state of it. The same logic runs the other way: formatting matters, but for the dull reason in the previous section, not because anything is judging your margins.
The score is a match report, not a verdict
The same resume and the same ad, run through a handful of free ATS scanners, come back with wildly different numbers for identical input. They cannot all be right at once, so most of them are wrong, and the reason is that they measure different things and call all of it an ATS score. A score only means something relative to one specific advert, and only if you can take it apart. That is also the honest test for any checker: can it show you what the number is made of, or is the low number the product? careerify's free check returns three numbers from a single model read, keyword match, semantic relevance and an AI-screener read, plus up to five of the ad's important terms that are missing from your resume. The big percentage on screen is only the plain average of those three, so the signal sits where they disagree: a low keyword match beside a high semantic score means you have the experience and named it differently, while all three low is usually not a wording problem at all. We do not simulate any named ATS product, and no tool, ours included, can back a claim about whether a particular employer's setup will pass you.
Only one of four reasons can be written away
Take the phantom robot out of the picture and four ordinary explanations cover most of the silence: a knockout question you answered outside a hard requirement, ranking below other applicants, a role that was never really open, and nobody opening your application at all. Knockout questions are the closest thing to genuine automatic rejection and still not what people picture, because they are questions a recruiter added to the form with answers that same recruiter marked as disqualifying, run against what you typed yourself. A rejection that lands within minutes therefore points at the form, not at the document you attached. Only the second explanation is a resume problem, and it is the one you can act on: search matches strings, so write system and standard names exactly as the ad writes them (SAP S/4HANA, Power BI, IFRS are too specific to guess at), put them in the skills section and in the bullets for your current role, and keep the local-language term beside the English one in Nordic ads that mix languages. The line is that every term needs a row of real experience behind it, because a recruiter picks a line and asks you to walk them through it, and you freeze on your own resume if AI wrote that line for you instead of with you.
Common questions
Does an ATS automatically reject your resume?
Rarely on its own. The system stores the application, parses it into fields, and lets a recruiter search, filter and rank. The automation that does exist sits in the form: screening questions a recruiter added, with answer options that same recruiter marked as disqualifying. The rule runs against what you typed yourself, and a person wrote it. No part of the chain reads your layout and forms an opinion about you.
Why was my application rejected within minutes?
Almost always a screening question on the form. Your answers there are structured data in known fields, so they can be evaluated the second you submit, while an attached document has no equivalent lever. The rejection email itself is a template that goes out when a recruiter moves a batch of applications to a rejected status, and some employers send instantly while others queue overnight. Speed tells you a rule ran, not that anything read your document. Reread the screening questions you answered, not your margins.
Can an ATS see white text on a resume?
Treat it as yes. Extraction pulls the characters out of the document, and colour and font size are instructions for drawing those characters rather than part of them. A white one-point paragraph is therefore invisible to your eye and completely unremarkable to whatever reads the file afterwards. The most common damage is human anyway: someone pastes the resume into a plain text field and the hidden paragraph shows up in black, in the middle of an application that was going fine.
Is 72 a good ATS score?
There is no universal pass mark, so the number on its own answers nothing. Ask what the 72 is made of instead. If the terms you are missing are ones you can honestly claim and have simply phrased your own way, the rewrite is quick and honest: call the thing what the ad calls it, because it is already true of you. If the number is low because the core requirement genuinely is not in your background, it is telling you something true, and rewording is the wrong tool.
Does resume formatting still matter?
Yes, but for a duller reason than the myth suggests. Multi-column layouts, text that is really an image and details buried in headers can land in the wrong field or never make it into the text at all, and then you do not appear in searches you should appear in. That is an absence rather than a rejection. One column, real selectable text and standard section headings are what you do about it, not to appease a filter but so the words you actually wrote end up in the fields a recruiter searches.
Want to see how your own CV holds up against one specific ad, start here: run an ATS check.