PDF or DOCX, what each one actually costs you, and what this product does with both
The choice is usually presented as a rule. It is a trade-off, and both sides of it are measurable on your own document in about a minute.
— the posting decides. where it does not, the reader does.
Two formats, one document, and a great deal of confident advice in both directions. The useful version of this decision is short: the posting decides, and where the posting is silent the difference is about who edits the file next. The résumé builder makes both from the same document and holds neither back.
What a PDF keeps
The layout, exactly. Typeface, spacing, page breaks, the column you chose. It is the format to send when the document is going to a person who will read it as you made it, and when you care that the second page starts where you decided it starts. It is also the harder one to alter, which is occasionally the point.
What a DOCX keeps
The text as text, in a container anybody can open and change. Recruitment agencies rebrand CVs, internal systems import them, and some application forms still ask for Word. Here the Word file is built entirely in your browser: the OOXML and the zip container are assembled in the page, with no network call at all during the export. That measurement is in your DOCX is built in your browser.
The PDF is rendered on our side, and nothing is kept
A PDF has to be rendered, so the document data is sent to our renderer, which uses headless Chromium. That route writes nothing to disk and nothing to the database; it returns a file. It is one of the short list of things that leave this browser at all, and the whole list is on why there is no account.
The measurement that complicates the advice
Our own round trip found something worth knowing, and then worth fixing. On 12 September a résumé pasted as text and the same résumé exported as a PDF and read back gave different counts on one content rule, because the renderer drew its bullets as layout rather than as characters, so the exported text layer held none. The rule that counts achievement lines saw seven in one case and four in the other, and the published score moved from 97 to 94. The exporter now writes the bullet marks into the text layer, and re-measured on 13 September both routes return 97 and both count seven lines. The whole experiment, before and after, is in pulling text back out of a PDF.
What that does not mean
It does not mean send a Word file to be safe. It means a rendered document is a different artefact from the text that made it, and the only way to know what yours extracts as is to check the one you are actually sending. The ATS résumé checker takes a PDF, a DOCX, a TXT file or a photograph, reads it in this browser first, and prints the words, lines and recognised headings it got out.
The one that is genuinely risky
A scanned or image-only PDF. There is no text layer to extract, and the parseability rule deducts heavily when the extracted text shows a broken-glyph pattern or has no readable section content. Scanned and image PDF résumés covers what to do with one.
Naming the file
Whatever the format, the file name is read by a human before the document is. Your name and the word résumé is enough, and résumé file name and format covers the conventions. The builder names the export after you rather than after a template.
The decision, in one line
Follow the posting. Where it is silent, send a PDF to a named person and a DOCX to an agency. Then check whichever one you are sending, because a format argument settled in the abstract is worth less than one measurement of your own file. Both exports are free with no watermark, and the ATS-friendly format guide covers the rest of what survives extraction.
One more thing worth doing once. Export both files from the same document and open each one. A DOCX opened in a word processor will reflow slightly, because the application makes its own decisions about fonts and spacing; a PDF will not. Seeing that difference on your own document settles the question faster than any argument about formats, and it takes two minutes.
Then keep the master in the builder rather than in a folder of files. The document in this browser is the thing you edit; the exports are copies taken for a particular application. That model is why a tailored version costs one export rather than one duplicated file, and it is worked through in one résumé or one per job.
Export both and compare
Build the document once and take either file. Nothing is held back and neither export carries a watermark.
Open the résumé builderRead the file format guideQuestions
- Which one should I send
- The one the posting asks for. Where it does not say, a PDF preserves what you designed and a DOCX lets an agency rebrand it. Both are free here and neither carries a watermark.
- Does the DOCX leave my browser
- No. The Word file and its zip container are built in the page itself, with no network call during the export. The PDF is different: it is rendered on our server through headless Chromium.
- Is the PDF stored anywhere
- No. The render route writes nothing to disk and nothing to the database. It returns the file and that is the end of it.
- Do parsers struggle with PDFs
- A well-made text PDF extracts cleanly. What does cause trouble is a scanned or image-only PDF, and there is a guide on those. Our own round-trip measurement found a subtler effect on how bullets are counted.
- Which format does the checker accept
- PDF, DOCX, TXT or a photograph, read in this browser first, and pasted text as well. The report prints what it managed to read.
Conxfolio is a free set of four career tools: a résumé builder with 37 rendered layouts, an ATS résumé checker that prints its own arithmetic, a cover letter builder that traces every proof paragraph back to the line it came from, and a portfolio builder with 20 authored designs. There is no account to create, nothing is held back for a paid plan, and no language model is used anywhere in the product, so the readers, the score and the letter are deterministic code you can check.