An applicant tracking system reads your resume before a person does. It looks for a name, a job title, dates, employers and skills, and it reads them in the order the file puts them. A template that hides those behind columns, text boxes or a graphic sidebar makes the system work harder than it will, and the result is a profile with holes in it.
The templates below are built the way parsers expect. One column for the content that matters, real headings instead of images, standard section names, dates in a consistent format, and no text locked inside a shape. They still look like something you would send to a person.
You fill everything in the browser, switch template without retyping anything, and see your ATS score and the list of what is missing free once you sign in. There are 99 templates. The finished resume, the templates and the PDF and Word download are Pro.
If you already have the job posting, paste it in. The AI lifts the requirements it actually asks for into your profile and skills, using only what you wrote yourself. It does not invent numbers, employers or dates.
An ATS is a database with an upload form in front of it. When your file arrives it runs a parser over it, and the parser tries to fill in fields: name, email, phone, current title, employer, start date, end date, education, skills. Whatever it cannot place is either dropped or dumped into a single unstructured blob that nobody searches.
Recruiters then work from those fields, not from your PDF. They filter by job title, by years in a skill, by location, by education level. A resume that parsed badly is not rejected, it is simply absent from the filter that would have found it. That is the failure mode worth designing against, and it is quieter than a rejection email.
The systems in common use are Workday, Greenhouse, Lever, SmartRecruiters, iCIMS, Taleo and Ashby. They differ in quality. Greenhouse and Lever parse well and forgive a lot. Taleo and older Workday configurations are stricter, and they are the ones large employers and public sector bodies tend to run. Build for the strict end and the rest look after themselves.
Almost every parsing problem traces back to a layout decision that looked good on screen. In rough order of how often it causes damage:
Order matters, because the parser and the recruiter both read from the top and the recruiter stops early. Contact block first, on the page itself and not in the header. Name, city and country, phone, email, and one link if it is relevant, usually LinkedIn or a portfolio.
Then a professional summary of three or four lines. Not an objective, and not adjectives. Job title, years, the two things you are strongest at, and one measurable result. This is the block the recruiter reads before deciding whether to read the rest, so it should say what the posting is asking for, in the posting’s own vocabulary where that is honest.
Then experience, most recent first, with employer, job title, location and dates on their own line. Under each role, three to five bullets that lead with a verb and end with an outcome. "Rebuilt the onboarding flow, which cut drop off from 38% to 22% in one quarter" beats "Responsible for onboarding" every time, because the first one survives a filter search and the second one describes a job description rather than a person.
Then education, then a skills block written as plain comma separated text rather than a graphic. Certifications, languages and projects go after that when they support the application, and are left out when they do not.
The posting is the keyword list. If it says Kubernetes, write Kubernetes and not container orchestration. If it says accounts payable, write accounts payable. Parsers and recruiter searches match on strings, and a synonym you preferred is a string that does not match.
Where a term has an abbreviation, use both once. "Search engine optimisation (SEO)" covers a recruiter who searched for either. Do the same for certifications, since some people search PMP and some search Project Management Professional.
What does not work: a block of keywords at the bottom, white text on a white background, or a skills list containing every tool you have ever heard of. The block reads as spam to the human who opens the file, the white text is detected by most current systems and it ends the application when it is found, and the inflated skills list gets discovered in the first ten minutes of the interview.
The honest version is simple. Claim what you have used, name it the way the posting names it, and put the strongest claims where they are read first.
Send a PDF unless the application form asks for something else. Every current applicant tracking system reads PDF text, and it holds the layout on every device. The old advice to always send .doc came from a period when parsers really did struggle with PDF, and it has been out of date for years.
Two exceptions. Some government and university portals still specify .doc or .docx, and a few recruitment agencies want Word so they can strip your details and add their own branding. When a form names a format, use that format.
Never send a PDF that is a scan or an export from a design tool as a flat image. It looks identical to a person and contains no text at all, so the parse comes back empty.
Name the file so it is findable in a folder of four hundred others: your surname, your first name, and the word Resume or CV. "Tarabya-Moustafa-Resume.pdf" is right. "resume-final-v3.pdf" is how you disappear.
A single main column, headings written as text rather than images, standard section names like Experience and Education, a consistent date format, and no content trapped in text boxes, tables or headers. Parsers read the file in order, so anything that breaks the order breaks the read.
No. After you sign in, the ATS score, the gap list and a blurred preview are free. The finished resume, all 99 templates and the PDF and Word download are Pro at $16.99 a month or $99 a year.
Modern systems handle more than they used to, and a two column layout will usually parse. The risk is not that it fails outright, it is that the reading order scrambles and your most recent job ends up next to a skill list. A single column removes the question.
Send a PDF unless the posting asks for something else. PDF keeps the layout on every device and every current applicant tracking system reads it. If a form only accepts .doc or .docx, use the Word download instead.
Yes, where they are true. If the posting asks for a tool you have used, name that tool the way the posting names it. Do not paste a keyword list or hide white text, both are obvious to a recruiter and some systems flag it.
One page for early career, two once you have a decade behind you. Length is not what the system scores. What matters is that the first third of the first page answers who you are and what you do.
Yes. Content and layout are stored separately, so switching template never means retyping. Try a few and keep the one that holds your most recent role and your top skills on the first page.
Not from us. You can fill any of the 99 templates in the browser and see your ATS score and gap list free after signing in, but the download is Pro. Free template files elsewhere are usually Word documents built on tables and text boxes, which is the layout that causes the parsing problems in the first place.
A normal template is designed to look good on screen. An ATS template is designed to be read correctly in order by software first and to look good second. In practice that means one main column, text headings, standard section names and no content inside shapes or tables.
Sign in free and see your ATS score. You only pay if you want the finished resume, the PDF and Word download, or all 99 templates.
Built by Moustafa Tarabya at DT Nova