The Formatting Mistakes That Break ATS Parsing

Before a recruiter spends a second on your resume, a piece of software usually reads it first. That software, an applicant tracking system, takes your document apart and tries to sort it into structured fields. When it succeeds, your experience lands in the right boxes and you get ranked fairly. When a formatting choice confuses it, whole sections can arrive garbled or vanish. The scale of this is not small. According to the World Economic Forum's Future of Jobs Report 2025, 86 percent of employers expect AI and information-processing technology to transform their business by 2030, and hiring is one of the first places that shift shows up. At Metaintro, where we track more than 50 million jobs across our platform for over 1.8 million users, the pattern we see is consistent. Strong candidates lose ground not because of what they did, but because of how their document was built. The World Economic Forum's companion press release on the 2025 report projects a net gain of 78 million jobs by 2030, so there is no shortage of roles to apply for, only a growing wall of software between you and them. This guide walks through the nine formatting mistakes that most often break ATS parsing, and gives you the exact fix for each.

Why Does A Small Formatting Choice Break An Entire Resume?

An applicant tracking system does not read a resume the way you do. It runs a parser, a program that scans the file top to bottom and tries to guess which piece of text belongs in which field. It looks for patterns. A line near the top that matches an email pattern goes into the contact field. A block of dates next to a company name becomes a job entry. A cluster of nouns under a heading called "Skills" becomes your skills list. The whole system depends on the text sitting in a predictable reading order and in a place the parser expects to find it.

That is why a small choice can cause an outsized failure. When you put your phone number in the document header, you are technically putting it in a layer that many parsers skip. When you arrange your work history in two columns, you may be inviting the parser to read straight across the page and merge two unrelated jobs into one line of nonsense. The information is all there when a human looks at the printed page. The parser just never got clean access to it.

The SHRM State of AI in HR 2026 report found that recruiting is the single most common area where organizations deploy AI, and that resume parsing is one of the top real-world uses. In other words, the parser is not a rare gatekeeper you might occasionally meet. For a large share of the roles we track in our postings at Metaintro, it is the first reader every single time. That makes clean parsing a baseline requirement, not a nice-to-have. If you want to see the flip side of these mistakes, the guide on why qualified candidates get rejected by resume bots explains how good applicants get filtered out by mechanics like these.

Why Do Headers And Footers Swallow Your Contact Details?

This is the most common and most damaging mistake, because it can cost you the one thing the recruiter needs most, a way to reach you. Word processors treat the header and footer as a separate layer from the body of the document. It is convenient for page numbers. It is dangerous for anything a parser needs to capture, because many applicant tracking systems read only the main text layer and never look inside the header or footer.

Picture a resume where the name, phone number, and email address sit centered in the document header, styled beautifully across the top of the page. To your eye it is the first thing you see. To a parser that skips the header layer, the contact field comes back empty. The system may still ingest your experience, but it has no way to tell a recruiter how to reach you, and some platforms will flag or deprioritize a record with no contact information.

Before: your phone number and email live inside the document header, so the parser reads a resume with no way to contact you.

After: the very same details sit on the first line of the body text, directly under your name, in the normal flow of the document, and the parser captures every one of them.

The fix takes thirty seconds. Delete the header and footer entirely and retype your name and contact line as ordinary body text at the top of the page. Do the same with any footer that holds your name or a page label. If you want a fuller walkthrough of where each element belongs, the companion guide on resume layout that AI parsers read cleanly covers placement in detail.

Why Do Text Boxes Make Your Words Disappear?

A text box is a floating container that sits on top of the page rather than in the natural flow of the text. Designers love them because you can drop one anywhere and move it around freely. Parsers hate them for exactly the same reason. Because a text box is anchored outside the main text stream, many systems read right past it as if the words inside were not there at all.

The trap is that text boxes are often used for the highest-value content. People put their professional summary in a shaded box at the top. They put a sidebar of core competencies in a box down the left. They frame a key achievement in a callout box for emphasis. When the parser ignores the box, it ignores your best material and keeps only the plain paragraphs around it. You end up ranked on the weakest half of your own resume.

The fix is to never use a text box for anything that must be read. Take the summary, the skills, the highlighted achievement, and type them as ordinary paragraphs and lists directly in the body. If you built your resume from a design-forward template, the boxes may be invisible until you click them, so check by selecting all the text and confirming every word highlights. Anything that does not highlight is trapped in a box. The guide on formats that survive ATS parsing explains which file types preserve this clean structure and which quietly break it.

Why Do Tables And Columns Scramble Your Experience?

Tables and multi-column layouts are the classic resume trap, and they fail in a way that is worse than simple deletion. Instead of dropping your content, they rearrange it into gibberish. A human eye reads a two-column layout the way the designer intended, down the left column, then down the right. A parser often reads straight across the page, left to right, one line at a time. When it does that to a two-column resume, it stitches together the first line of your left column with the first line of your right column, then the second line of each, and so on. Your carefully organized page comes out as a jumble of half-sentences.

Tables cause a related problem. Even a single-row table used to line up a job title on the left and a date on the right can confuse a parser into misreading which date belongs to which role, or into treating the whole table as one undifferentiated block. Skills grids are the worst offender, because they pack many short items into a tight lattice of cells that the parser cannot reliably unpack.

Before: your skills sit in a three-by-four table, twelve cells of short phrases, and the parser reads across the rows so "project management" and "budgeting" and "Spanish" merge into one scrambled string that matches nothing.

After: the very same skills run as a single plain line of body text, Project management, Budgeting, Data analysis, Spanish, Stakeholder communication, read cleanly left to right with no cells to confuse the reading order.

The fix is to remove every table and every column and rebuild your resume as one single running column that flows top to bottom. List your skills as a simple comma-separated line or as plain bullets, one idea per bullet. For the money-side of this, keyword placement without the grid, see resume keywords without keyword stuffing.

Why Do Images And Logos Come Back Empty?

An applicant tracking system reads text. It does not read pictures. Anything you place on your resume as an image, a company logo, a headshot, a graphic skills bar, a chart of your metrics, is invisible to the parser, because there is no text for it to extract. Worse, some systems will choke on a heavily graphical file and either reject it or return a mostly blank record.

The most common version of this mistake is the visual skills meter, the little bar or set of dots that shows you at "90 percent" in one tool and "70 percent" in another. To a human it looks polished. To a parser it is a picture with no words, so your skills section comes back empty, and the number was meaningless anyway because you assigned it to yourself. Company logos beside each job are another version. If the company name exists only inside the logo image, the parser never learns where you worked.

The fix is to make sure every fact on your resume exists as real, selectable text somewhere on the page. Type each employer's name as text next to the role, even if you also want a logo for a human reader, though the safest move is to drop the logos entirely. Replace visual skill bars with plain text, and if you want to signal proficiency, say it in words like "advanced" or "conversational." A photo of yourself adds nothing a parser can use and can trigger bias screens, so leave it off. For a deeper look at how the reading step works under the hood, see how AI resume screening works.

The stakes are higher now that AI sits on both sides of the hiring process, screening resumes on one side and reshaping the roles you apply for on the other. Lacey Kaelani, CEO of Metaintro, told People Managing People that "AI is not completely eliminating roles, but instead restructuring roles and therefore slowing hiring for some jobs." When fewer roles open and each one draws more applicants, a clean parse is what keeps you in the pool long enough to be seen.

Why Do Icons Erase Your Contact Information?

Icons feel like a harmless design flourish, but they cause a specific and painful failure. When you replace the word "Phone" with a little phone symbol, or the word "Email" with an envelope glyph, you are handing the parser a picture where a label should be. Many systems rely on nearby text cues to know what a field is. Strip out the words and the parser may not recognize that the number beside the phone icon is a phone number, or that the string beside the envelope is an email.

The result is the same category of damage as the header mistake. Your contact details are on the page, a human sees them instantly, and the parser files them as unlabeled noise or drops them. Location pins, calendar icons for dates, and link icons for your portfolio all create the same gap between what a person understands and what the machine can extract.

Before: an envelope icon sits next to your email and a phone glyph next to your number, with no words, so the parser sees two floating strings and cannot tell which is which.

After: the plain labels return, Email followed by your address and Phone followed by your number, all as text on one line, and the parser tags each field correctly.

The fix is to write the label out in words. Use "Email," "Phone," and "Portfolio" as plain text, and if you love the look of icons, know that the tradeoff is a real risk to the most important line on your resume. Our own breakdown on the blog, Resume Icons in 2025: When They Help Your Application and When They Hurt, walks through where a stray icon helps a human reader and where it quietly costs you the parse.

Why Do Special Characters And Symbols Confuse The Parser?

Not every character on your keyboard survives the trip through a parser. Decorative bullet symbols, fancy dividers, arrows, checkmarks, star ratings, and unusual dash and quote marks can all come back as garbage characters or get stripped in a way that fuses two words together. When a special character sits between two pieces of information, its failure can merge them, turning "Manager, Operations" into one unreadable token that matches nothing a recruiter searched for.

Curly, stylized quotation marks and long dashes are a quieter version of the same issue. They usually render, but on some systems they convert to placeholder symbols that make your text look broken and can interrupt the parser mid-line. The safest characters are the ones every system agrees on, the standard letters, numbers, a plain hyphen, a comma, a period, and a simple round bullet.

Before: your title reads "Senior Analyst to Finance" with a fancy arrow glyph between the words, and the parser turns the arrow into a stray symbol that welds the two words into one unsearchable string.

After: the same line reads "Senior Analyst, Finance" with a plain comma, and the parser cleanly separates the role from the department.

The fix is to strip your resume down to standard characters. Use a plain round bullet for lists, a regular hyphen where you need a dash, and straight quotation marks. Avoid emoji, star ratings, and decorative symbols entirely. If you copied text from a designed template or a web page, retype it or paste it as plain text first, because hidden formatting characters often ride along invisibly. To confirm none slipped through, run your file through the check in test your resume against an ATS.

Why Do Nonstandard Fonts And Odd Date Formats Trip The Reader?

Two smaller mistakes deserve a section together, because both come from choices that look fine on screen and fail underneath. The first is fonts. When you embed an unusual or decorative font, some systems cannot map its characters correctly, and letters come back substituted or dropped. Very thin display fonts and script fonts are the biggest risk. The safe move is a common, widely supported font in a normal weight, so the parser reads exactly the letters you typed. Keep the size in the ordinary range, roughly 10 to 12 point, so nothing gets treated as an accidental heading or ignored as fine print.

The second is dates. Parsers are trained to recognize employment dates in familiar patterns, and they use those dates to build your work-history timeline and to calculate years of experience. When you write dates in a nonstandard way, an ambiguous all-numeric format, a season instead of a month, or just a start year with no end, the parser can misread your tenure or drop the entry from the timeline. That can quietly knock you below a "years of experience" threshold you actually clear.

Before: a job shows "twenty-two to present" in a stylized font, and the parser cannot tell whether that means 2022 or something else, so it leaves the role off your calculated timeline.

After: the same job reads "March 2022 to Present" in a standard font, and the parser places it correctly and counts the full tenure.

The fix is to write every date as a spelled or standard month and a four-digit year, like "June 2021 to March 2024," and to use "Present" for a current role. Keep the format identical across every entry so the parser sees one consistent pattern. For the full positive build, start-to-finish, use the how to write an ATS-friendly resume guide, and if you are reusing a design file, check it against resume templates and ATS before you trust it.

How Do You Make Sure The Machine Reads The Real You?

Every mistake in this guide shares one root cause. A formatting choice that looks good to a person hides your information from the software that reads first. The way to protect yourself is to build every resume twice in your head, once for the human who will judge your experience and once for the parser that decides whether the human ever sees it. The good news is that the parser-friendly version is not ugly. A clean, single-column, plain-text resume with real headings and standard characters reads well to both audiences.

Start with a simple habit. After you finish a resume, select all the text and confirm every word highlights, which proves nothing is trapped in a box or an image. Then save a plain-text copy of the file and read it top to bottom. That stripped version is close to what the parser sees, and if it reads in the right order with your contact details, jobs, dates, and skills all present, you have cleared the biggest hurdles. This matters more every year. The PwC 2026 Global AI Jobs Barometer found a 62 percent wage premium for workers with AI skills, and the same forces pushing AI into every job are pushing it into every hiring pipeline. Gallup reported that 45 percent of employees used AI at work as of the third quarter of 2025, and analysis in the McKinsey Global Institute report on agents and automation found AI fluency in the workforce grew roughly sevenfold, from about 1 million people in 2023 to about 7 million in 2025. Independent tracking agrees the shift is broad. The Pew Research Center found that about 21 percent of United States workers used AI in their job in 2025, up from 16 percent a year earlier, and the Stanford AI Index 2025 documents the same steep climb in real-world adoption, and the OECD reports that member countries may be on the cusp of an AI revolution in the workplace. The reader on the other side of your resume is increasingly a machine, so the reliable move is to build a document it can read.

At Metaintro, what we have seen across the roles we track is that candidates who fix these nine mistakes stop losing interviews they were qualified for. It is the cheapest improvement in a job search, because it costs no new skills and no new experience, only a cleaner document. Once your formatting is clean, the next move is matching your content to each posting, which the guide on matching resume keywords to each job with AI covers, and if you are moving between fields, getting past AI screening when changing careers shows how to keep the parser on your side when your titles do not line up.

Related Articles

• How to Write an ATS-Friendly Resume in 2026
• AI Resume Screeners Now Prefer AI-Written Resumes
• Half of Job Seekers Want ATS Banned
• Action Verbs That Strengthen Your Resume
• The Best Skills to Put on Your Resume
• Resume Icons in 2025: When They Help and When They Hurt
• How to Write a Resume That Beats AI Screening in 2026
• How AI Hiring Tools Can Quietly Screen Out Qualified Candidates
• The Ultimate Resume Writing Guide
• The Cover Letter That Actually Gets Read

People Also Asked

Q: Will an applicant tracking system really reject my resume just for a table?

A: It rarely sends a flat rejection for a table alone, but it can scramble the content inside the table so badly that your resume ranks far below where your qualifications should place you, which has the same effect. The safe move is to rebuild any table or two-column section as one single running column of plain text. The SHRM State of AI in HR 2026 report confirms recruiting is the top area where organizations deploy AI, so assume a parser reads first and give it clean structure.

Q: How do I know if my formatting is breaking the parse?

A: Do two quick tests today. First, open your resume, select all the text, and check that every word highlights, since anything that does not highlight is trapped in a text box or an image the parser cannot read. Second, save a plain-text copy and read it top to bottom, because that stripped version is close to what the software extracts. If your contact line, jobs, dates, and skills all survive in the right order, your formatting is clean.

Q: Are PDFs safe for an applicant tracking system?

A: A simple, text-based PDF built from a clean single-column layout usually parses well, but a PDF exported from a heavily designed template can hide text inside images or boxes that the parser skips. When a posting lets you choose, a plain text-based document is the lowest-risk option. Check whichever file you submit against the steps in the guide on resume formats that survive ATS parsing.

Make your resume work for you. Metaintro shows how real roles get posted and screened, so you can tailor your resume to clear the bots and reach a human. Sign up to match your resume to live openings and see what recruiters actually screen for.