ATS Parsing: How Greenhouse, Lever and Workday Read Your Resume

An applicant tracking system, or ATS, stores job applications. Parsing is how it turns your file into fields it can use. Those fields usually include your name, contact details, job titles, employers, dates, school and skills.

The software that does this work is called an ATS parser. It reads the text in your file. Then it sorts that text into the right boxes on your profile.

You are likely to meet one of these systems. Jobscan's 2026 report found a detectable ATS at 487 of the 500 Fortune 500 companies, or 97.4%. Over 40% of those companies used Workday. In Jobscan's wider data from 13,130 companies, the most common systems were these:

A parse error does not mean you are rejected. It means the system could not fill in your profile on its own. Greenhouse calls this an unsuccessful resume parse. The resume is only attached to the candidate, and a person has to type the details in by hand.

That extra work matters because recruiters often search by skills. Jobscan's State of the Job Search report says 76.4% of surveyed recruiters look for and sort candidates using skills from the job posting. A profile with missing or wrong fields gives those searches less to find.

Layout trouble was not common in our own data. Of 2,225 detailed resume checks, 96.2% passed JobResumeAI's formatting safety check and 3.8% did not. The same study found a median keyword match of 25%. So parsing is one step, and matching the job is another. We cover the second step later in this guide.

Parsers differ from one product to the next. The steps below describe the general path. They are not the code of any one system.

1. 1

   It reads the text in the file

   The parser pulls the words out of your PDF or DOCX. A picture of text has no words to pull, so there is nothing to read.

2. 2

   It finds the sections

   It looks for headings such as Experience, Education and Skills. Clear headings help it split the page in the right places.

3. 3

   It looks for details

   It looks for a name, an email, a phone number, job titles, employer names, dates and schools.

4. 4

   It fills the profile

   It puts what it found into fields on the candidate profile. Lever lists name, organization and contact details as examples.

5. 5

   It keeps your file

   Your original file stays with the profile. Lever keeps resumes that come with an application in a resume area on the candidate profile. Greenhouse attaches the file even when parsing fails.

How an ATS parser reads a resume

Greenhouse lists the resume problems that can cause a failed or partial parse. The list is a useful guide to ATS resume layout, even if you apply through another system.

The Greenhouse page is written for recruiters. It describes how the parser behaves. It is not a promise about every application form. Treat it as a strong hint, not a rule.

ATS resume layout and parse risk

People often ask about these items one at a time. The short answer is that some systems cope and some do not. You cannot know which system an employer uses, so the lower risk choice is the same each time.

Start with the application. If it asks for PDF, send PDF. If it asks for DOCX, send DOCX. The employer's instruction matters more than general advice about resume file format.

If either is accepted, the file type matters less than what is inside it. A PDF saved from Word or Google Docs holds real text. A scan or a photo of a page does not. Lever suggests a simple test. Try to select the text with your cursor.

So, should a resume be PDF or Word? Greenhouse and Lever both accept both. Choose the version that stays clean after export, and check it with the steps later in this guide.

File rules for Greenhouse, Lever and Workday

The three systems share the same basic idea. Their help pages and setups differ. Below is what each source says, and where we could not find an answer.

Think of this as a home resume parser test. It will not copy any one system. It can still show the problems that vendors name. Use your final file.

1. 1

   Open the final file

   Use the PDF or DOCX you plan to send. Do not test an older draft.

2. 2

   Try to highlight text

   In a PDF, drag your cursor across a few lines. If you cannot select the words, the file may be an image.

3. 3

   Copy and paste into a plain text editor

   Select all, copy, and paste into Notepad or a plain text window. This strips the design and shows the order of the text.

4. 4

   Read the pasted text from top to bottom

   Look for your name first, then contact details, then sections in order. Watch for broken lines, missing words and odd symbols.

5. 5

   Check the file size

   Keep the file well under 2.5 MB. Large images are the usual reason a resume is big.

6. 6

   Compare the resume with the job

   Run the resume against the job description and review the gaps. Add a term only if it is true for you.

Answer each question about your file. The checker lists what to fix first. It runs in your browser and does not upload your resume.

It is a self check based on problems that vendors document. It cannot tell you how a specific employer's system will read your file.

Print the checklist while you prepare your file. Open the CSV in a spreadsheet and fill it in after you upload. The copy lines below show what parse friendly text can look like.

Start with the changes that Greenhouse names, because they are the most likely to matter.

Move your name and contact details out of the header and into the body. Switch to one column. Replace tables with plain lines. Turn text boxes into normal paragraphs. Remove photos and word art. Retype any spaced out headings in normal text.

Then check the details a parser looks for. Write job titles in full. Write company names as the employer writes them, including Inc. or LLC when it is part of the name. Use real details instead of placeholder text.

Last, export the file again from your editor. Save a PDF from Word or Google Docs instead of scanning a printed page. Check the size. If the file is near 2.5 MB, remove large images.

A clean parse gets your details into the system. It does not show that your resume fits the job. The next step is to compare your wording with the job description.

In our study of 2,225 detailed checks, 96.5% had at least one term that JobResumeAI did not find in the resume. The median keyword match was 25%. A missing term can point to a wording gap or to a real gap in experience. Add a term only when it is true for you.

Try the quick scan below on a job posting. Use the result as a starting point, not as a list to copy.

We want to be clear about what we know and what we do not.

• Vendor help pages describe how a parser works in general. Employers can set up their own forms, so your result can differ.
• The Greenhouse page we cite is written for recruiters. It is not a promise about each application form.
• We could not find one public Workday page with file rules for applicants.
• Jobscan's numbers come from career page scans and scans by its own users. They show how common each system is, not how any one employer uses it.
• JobResumeAI's numbers come from a self selected set of resume checks. They are diagnostic flags, not employer hiring results.
• Vendors update their help pages. We checked them in October 2026.

• Did you use the file type the application asks for?
• Can you highlight the text in your PDF with your cursor?
• Is the file well under 2.5 MB?
• Are your name, phone and email in the main body of page one?
• Does the resume use one column, with plain headings?
• Are job titles and company names written in full?
• Did the copy and paste test read cleanly from top to bottom?
• After you upload, did you check each filled in field?
• Is every detail in the file true and yours?