An applicant tracking system does not score your resume and reject it. It reads the file into structured fields, stores the result, and makes it searchable for a recruiter. Almost everything that goes wrong at this step is a formatting failure, not a judgement, and almost all of it is fixable in an afternoon.
That distinction matters because the folklore around these systems has convinced a lot of capable people that an invisible algorithm is throwing their applications away. The truth is duller and much more actionable.
What Does an ATS Do?
An applicant tracking system is a database with a workflow attached. When you submit an application it does three things. It ingests your file and pulls the text out. It maps that text onto the fields it understands, which are name, contact details, employers, job titles, dates, education and skills. Then it files the record under a pipeline stage so a recruiter can see where you are.
None of that involves an opinion about you. The software is doing clerical work at speed.
The reason the myth persists is that a small number of systems can be configured to filter, and some employers do configure them. But the filters that exist in practice are blunt and legal rather than subtle and evaluative. Do you have the right to work in this country, do you hold the licence this role legally requires, will you relocate. Those are knock-out questions the employer wrote, usually presented to you as questions during the application. They are not a hidden quality score.
The software is not deciding whether you are good. It is deciding whether your file can be read.
Where Do Resumes Fail?
Resume parsing works on two things, the text layer of your document and the order that text appears in. Every common failure comes from one of those being different from what your eyes see.
Two columns. A layout with a narrow sidebar looks organised to a human. A parser reading in document order can interleave the two, so your skills list arrives spliced line by line into your work history. The result is unreadable, and it is your document that looks incoherent.
Tables. Dates in one cell and titles in another often flatten into a single run of text with no relationship preserved. Your three most recent roles can end up looking like one job with three names.
Headers and footers. Some parsers skip them entirely. If your name, email and phone number live in the header, you can submit an application that arrives with no contact details attached. This one is quietly common and completely fatal.
Text inside images. A logo, a skills chart, an icon with your job title baked into it, all invisible. If information exists only as pixels, it does not exist.
The wrong kind of PDF. A PDF exported from a word processor carries a real text layer. A PDF that is a scan, or one exported from a design tool as flattened artwork, may carry none. Everything else can be perfect and the file will still arrive empty.
The Thirty-Second Test
Open your resume. Select all. Copy. Paste into a plain text editor.
What appears is close to what a parser sees. Read it in that form and the problems announce themselves. Sections in the wrong order, columns interleaved, a job title missing, your phone number absent because it was in the header. If the plain text version reads sensibly top to bottom, your formatting is fine. If it does not, no amount of keyword tuning will save it.
This test is worth more than every ATS-checker tool on the internet, and it costs nothing.
What Are Keywords For?
Here is the part the advice industry gets backwards. Keyword matching does matter, but not because a robot is scoring you. It matters because of boolean search.
A recruiter with four hundred applications does not read four hundred applications. They type a query such as React AND TypeScript NOT contract. Whoever appears in that result set gets read. Whoever does not, does not, no matter how well qualified.
So the goal is not to hit a keyword density target. The goal is to be findable by a human running a plausible query. That means naming the things you can genuinely do, in the words your industry actually uses for them. If the job description says "TypeScript" and your resume says "TS", you may not appear. If you have done accessibility work and never used the word "accessibility", you will not appear in a search for it.
Three things do not work, and the first two will actively cost you. White text on a white background, invisible keyword blocks, or a "skills" section listing forty technologies you have touched once. The first two are visible to any human who opens the file and read as an attempt to cheat. The third dilutes the terms that are actually true.
What Should You Do?
Keep the layout in one column. Use ordinary section headings that say what they are, such as Experience, Education and Skills. Put your contact details in the body of the document, not in a header. Export a PDF with a real text layer, and check it with the copy-paste test. Then make sure the words for the things you can do appear somewhere true on the page.
That is the whole of ATS-friendly formatting. It is not a dark art, and it does not require a special template.
The reason to get it right is not that the machine is your enemy. It is that the machine is the only route to the human, and a file the machine cannot read never reaches them. Once it does, everything that decides the outcome is a person, first a recruiter scanning for a reason to keep reading, then a hiring manager deciding whether you can do the job.
Which is why the next question worth asking is what happens to the file after that. We wrote about what happens to your application after you hit send, and about why one carefully prepared application beats fifty generic ones.
