Resume PDF spacing, line breaks and paragraphs in text extraction

What spacing did to the extracted text

Five spacing cases against a control with ordinary headings.

Letter-spaced headings

Tracking out a heading (CSS letter-spacing, Word's Expanded character spacing, or spaces typed between letters) widens the gap between glyphs until it looks like a word break. pdf.js and our checker took the bait in both fixtures and returned single letters separated by spaces. Our checker then failed to recognise S K I L L S as a heading and returned no skills. We fixed that on 20 October 2026 after these tests: a line of single letters with spaces between them is now matched as the heading it spells, so both fixtures give 5 skills. The text it shows still has the spaces, because that is what the PDF holds.

PDF 32000 §14.8.2.5 (checked 19 October 2026) asks tagged PDFs to include "the spacing characters that would be present to separate words in a pure text representation", so that a reader "does not need to guess about word breaks based on information such as glyph positioning on the page". Untagged or not, a letter-spaced heading gives the reader either real spaces or gaps that look like them. pdftotext rejoined both versions, including the one with typed spaces, which shows how much the result depends on each extractor's own word-gap rules.

Greenhouse names the problem directly (checked 19 October 2026): "A resume with spaces between the letters. While it may appear cohesive to the naked eye, the parser won't recognize the separate letters as a single word and can't make sense of the data."

Justified text

Justification stretches the spaces between words, never inside them, so every reader still found the word boundaries. Our narrow justified summary extracted word for word in pdf.js, pdftotext and our checker. Justify if you like how it looks; it isn't an extraction risk on its own.

Line breaks and paragraphs

A PDF has no paragraphs, only positioned lines. A bullet that wraps at the margin and a bullet with a hard line break typed into it look the same on the page, and all three readers returned both as two lines. pdf.js documents that getTextContent hands back text items with whitespace normalised; pdftotext marks a block break with a blank line. Deciding which lines belong together is left to the parser.

Our checker rejoined each pair into one bullet, filing 2 bullets. In a DOCX the difference is real: a wrapped bullet is one paragraph and one line, while a hard break (Shift+Enter in Word) splits it in two inside the same paragraph.

How to fix spacing in a resume

• Set headings with normal letter spacing. Use weight, size or a rule for emphasis instead.
• Never type spaces between letters to stretch a word.
• Let bullets wrap on their own. Don't press Shift+Enter to control where a line breaks.
• Keep one idea per bullet so a wrapped line can't be mistaken for a new one.
• Justified or left-aligned both extract; choose on looks.

How we tested

We built every fixture from the same fictional one-page resume, changing one thing at a time. PDFs come from Chromium 153.0.8010.12 (printed through Playwright 1.63.0) and from LibreOffice 26.8.1.1 converting our DOCX files, the way Word users export. Each file was read three ways: pdf.js 6.3.289 getTextContent, poppler pdftotext 26.10.0 in its default mode, and the importer the ATSPrep checker runs in your browser. DOCX files have no PDF text layer, so only our importer reads them. The code blocks on this page are the readers' output, copied by script; private-use and control characters are written as [U+XXXX] because they would otherwise be invisible.

These are three common open readers, not any employer's applicant tracking system. Commercial parsers use their own extractors, so treat a failure here as a risk to remove, not a prediction of what a specific system will do.

Common questions