ATS-friendly resume formatting

The short answer

An ATS-friendly resume is one whose text comes out of the file in the right order, with nothing important locked inside a picture. In practice that means one main column, standard section headings, real text instead of graphics, contact details in the body of the page, and a PDF or Word file exported from an editor rather than scanned.

It does not mean a plain or ugly document. Color, a clear typeface, bold job titles and generous spacing all survive parsing. What does not survive is meaning that only the layout carries: a skill rating drawn as dots, a timeline drawn as a graphic, a sidebar whose text comes out in the middle of your job history.

The rest of this guide explains why, walks through one fictional resume before and after, and gives you a test you can run on any file in two minutes.

How a parser reads your file

A parser does not see your page. It sees a stream of characters in the order the file stores them, and then guesses which part is your name, which lines are job titles and which are dates.

In most simple layouts the stored order matches the reading order: top to bottom, left to right. Problems start when the two differ. Columns, text boxes, tables and floating shapes are the usual causes. You see a tidy page. The parser receives your skills list in the middle of your first job.

The second problem is text that is not text. A scanned page, a heading exported as an image, a word replaced by an icon. To a parser, that part of the page is blank.

Layout principles

Single-column main content is the safest default. Sidebar designs can work if the file keeps a logical text order, and the only way to know is to test the export, as described below.

Use standard headings: Summary or Profile, Experience, Education, Skills, Languages, Certifications. Creative titles can confuse parsers and humans alike.

Put your name and contact details at the top of the page body, not only in the document header or footer.
Give each role one clear line: job title, employer, location if relevant, and dates.
Write dates in one format throughout, for example "Mar 2022 – Jun 2024", and keep them next to the role they belong to.
Use ordinary bullets for achievements and keep each bullet to one idea.
If you use a second column, keep it for short items such as contact details and languages, never for your experience.

Typography and spacing

Stick to common fonts, sans-serif or serif. Decorative fonts sometimes extract as the wrong characters, so a name that looks elegant can arrive as a row of symbols. Avoid type smaller than about 10 points. Machines mostly care that characters are selectable text. Humans care about white space, so leave some.

Bold job titles and company names so a reader can scan the page.
Use bullets for achievements, not long paragraphs, in the experience section.
Write links out as full addresses, or as clear link text with the address visible. Plain-text extraction usually keeps the words and drops the link behind them.
Keep color for headings and accents. Never use it to carry meaning on its own, such as green for skills you are strong in.

File format

PDF is the usual choice for applications, because it looks the same on every screen. Word files are fine when an employer asks for them. Send a clean export without text boxes. Whatever the format, follow the employer's instructions when they give any.

Export from the editor you wrote in rather than scanning a printout. A scan is a photograph of text, and most parsers cannot read it. Name the file after yourself, such as "Firstname-Lastname-Resume.pdf", so the person who downloads it can find it again.

Some portals ask you to paste your resume into a plain text box. Keep a plain-text version for that, built from the same source, rather than pasting from the PDF and fixing the line breaks by hand each time.

A worked example: before and after

This example is fictional. Tomas is a data analyst applying for his next role. His resume comes from a downloaded design template and looks good on screen: his name in a colored banner, a dark sidebar with icons and skill bars, and his work history in a two-column table, dates on the left and job details on the right.

He runs the copy-and-paste test described below. Here is what the plain text shows.

His name is missing. The banner was exported as an image, so the text starts with the word "Experience".
His phone number is missing. It sits in the document header, which the text he copied did not include.
"SQL", "Python" and "Tableau" appear as three bare words. The bars that rated them are gone, and so is any sense of how he used them.
The table comes out as a block of four date ranges followed by a block of four job titles, so nothing shows which dates belong to which job.
His portfolio appears as the single word "Portfolio", with no address.

What Tomas changed

None of these problems is about his experience, and none needed new content. He changed the layout and kept everything he had written.

His name is now live text at the top of the page, in larger type and the same accent color.
A contact line sits under the name, in the page body: city, email, phone, and the portfolio address written out.
Each role is a heading line with title, employer and dates, followed by its bullets. The table is gone.
Skills are a plain list under a "Skills" heading. The bars are gone, and each tool is also named in the bullet where he used it.
He exports a PDF from the editor and runs the test again. The text now reads from top to bottom in the order a person would read it.

What the example shows

The new version still looks designed. It has the same font, the same accent color and the same content. The difference is that the information now lives in the text, not in the layout. That also helps the recruiter who reads it on a phone, or prints it, or opens it in a system preview that ignores the styling.

What usually goes wrong

Most parsing problems come from a short list of design choices. Each is easy to fix once you know it is there.

Skill bars, star ratings and charts. They carry no text, and the rating itself is a claim no one can check.
Important details in the header or footer. Some parsers skip those areas entirely.
Text boxes and floating shapes. Their text is often stored out of order, or apart from the rest of the page.
Tables used for layout. Cells can be read row by row, column by column, or not at all.
A scanned or image-only PDF. It may look like a resume, but to a parser it is an empty page.
Experience in a narrow column. If the main content sits in the sidebar, it is the part most likely to be scrambled.
Section titles that try to be clever. "My toolbox" is a skills section only to a reader who already knows it.
Link words without addresses. "Portfolio" or "GitHub" as clickable words loses the address in plain text.

The two-minute test

You do not need special software to check a file. Open your final PDF, select all the text, copy it and paste it into a plain text editor. Then read it from top to bottom and check five things.

Your name appears in the first lines.
Your email and phone number are there in full.
The section headings appear in the order you expect.
Each job title sits next to its employer and dates.
There are no strange characters where letters should be.

Two more quick checks

Search the PDF for your own email address. If the viewer cannot find it, that part of the page is an image, not text.

If a job portal prefills its form from your resume, read the prefill before you correct it. It shows you what that employer's parser took from your file, which is the most direct evidence you will get.

How Verlane handles it

Verlane builds the resume as structured text first and draws the layout from it, so the content never depends on a picture.

There are three templates. Minimal is a single column. Clean adds a colored sidebar for contacts, skills and languages. Vivid uses a colored header block above a narrow left column.
Every template downloads as PDF or Word, and both files are drawn from the same layout as the on-screen preview.
The free health check includes an ATS file readability check. It reads your original upload the simple way a basic parser would, with no layout repair, and reports whether your name, email, phone and the Experience, Education and Skills headings come through.
It runs the same read on Verlane's own PDF and Word downloads. Clean and Vivid use a side column, so you can check that the download holds up rather than take our word for it. Minimal is the safest choice when a portal behaves oddly.
In the editor you can rename, reorder and hide sections. Hidden sections are left out of the download and out of the match score.
If your only copy of your resume is a scan, Verlane can still read it when you upload it, and you can download a text-based file from the result.