Why Your Resume Does Not Parse

Parsing failures are not random. There are about six of them, they are all visual choices, and each one has a tell you can check in under a minute.

From the team behind PassATS. Free to read, no email required.

Two columns

What you see: skills down the left, experience down the right.

What the parser sees: text read left to right across both columns, so your job titles interleave with your skill list. "Senior Engineer Python 2021 to 2024 SQL Led migration Docker".

How to check: select all, copy, paste into a plain text editor. If the result is scrambled, so is the machine's copy.

Tables and text boxes

What you see: a tidy grid of skills, or a highlighted sidebar.

What the parser sees: cell fragments with no reading order. Text boxes are worse — many parsers skip them entirely, so that content simply does not exist.

How to check: the same copy-paste test. Missing content is the tell.

Contact details in the header or footer

What you see: your name and email neatly at the top of every page.

What the parser sees: in some systems, nothing. Header and footer regions are commonly ignored.

The consequence: you can be the strongest candidate in the pile and still be unreachable. Put contact details in the body of the first page.

Creative section headings

What you see: "Where I Have Made An Impact".

What the parser sees: an unclassifiable block. It looks for the words it knows — Experience, Education, Skills, Summary — and assigns everything underneath to that section. An unrecognised heading means the content beneath it never gets categorised.

This is the most common self-inflicted wound in resume design, because it feels like personality rather than a technical choice.

Words inside graphics

What you see: a skills wheel, a proficiency chart, a logo containing your name.

What the parser sees: nothing. An image is not text.

If a piece of information matters, it has to exist as selectable characters somewhere on the page.

Inconsistent dates

What you see: Jan 2023 in one role and 01/2023 in another.

What the parser sees: a history it cannot reliably order, and sometimes a gap that is not really there.

Pick one format and use it everywhere.