A resume PDF is refused when it is larger than the limit set by that employer's application system, and the limit differs from one system to the next. Shrinking it in the browser helps only if the file is large because of loose internal structure, not because every page is a scanned image.
The compressor here is a lossless structural rewrite built on qpdf 11.7.0 compiled to WebAssembly, and it runs inside the page, so the file is never uploaded. It repacks objects into compressed streams, rebuilds the cross reference tables and leaves image pixels untouched. Text heavy resumes exported from Word or Google Docs often lose a useful slice of their size. A resume built as one big image per page usually stays the same size or grows slightly, and the tool reports that honestly instead of claiming a win.
A job form does not say which part of the resume is heavy. You select the file, watch a progress bar stall, and get an upload error with no breakdown of where the bytes went. The file may be a clean two page text document that a loose exporter padded, or a scan in which every page is a photograph.
The two cases look identical in a file manager and behave differently under compression. Telling them apart decides whether the tool below can help at all.
Compress PDF
Rewrite a resume's internal structure to reclaim wasted bytes, with the real before and after sizes shown. The file never leaves your device.
Why a resume is rejected before it is uploaded
Applicant tracking systems set their own per file ceiling. That ceiling is not a public standard: it is a setting inside each system, it varies between vendors, and it can vary between two employers running the same vendor on different plans. A number copied from a help article is a number for somebody else's account, which is why this page does not quote one.
Where the check happens matters more. The form reads the size of the file the moment you select it, in your own browser, before the bytes leave your machine. When the answer is no, nothing was transmitted, and retrying the same file produces the same refusal.
What the compressor actually does to a resume
The tool is qpdf 11.7.0 compiled to WebAssembly. It loads into the tab, reads the PDF from disk into memory, rewrites it, and hands back a new file. Nothing is transmitted, and the 50 MB ceiling is enforced in the page, so a larger file is refused on the spot rather than uploaded and then rejected.
The rewrite is structural. Objects are packed into compressed object streams, streams the authoring app left uncompressed are deflated, and the cross reference tables are rebuilt. Standard mode does this at compression level 6. Maximum mode does it at level 9 and then linearizes the file, which rearranges it so a reader can draw page one before the rest of the document arrives.
What does not happen is image re-encoding. Every photograph, every scanned page and every screenshot inside the document is copied byte for byte. That one fact decides the outcome more than the mode you pick.
| What the resume is made of | What the rewrite does | What to expect |
|---|---|---|
| Text, embedded fonts, vector rules | Repacks and deflates the structures around the text | Often a real reduction, with pages that look identical |
| Scanned pages, one image per page | Copies each image untouched | Little or no reduction, sometimes slightly larger |
| Text plus a large photo or portfolio image | Shrinks the text layer and leaves the image alone | A small gain that the image quickly dominates |
There is also the case nobody advertises. A PDF that is already tightly packed can come out slightly larger, because object streams and a rebuilt cross reference table are not free. The tool reports the true sizes either way; if the download is bigger, keep the original.
Standard or Maximum
Maximum sounds like the stronger setting, and on compression level it is. The trap is linearization. To let a reader show page one immediately, qpdf writes a page index at the front and stores a second copy of some objects. On a short resume that overhead can outweigh the extra deflate pass, which is why Maximum is not reliably the smaller output.
Both modes are free and unlimited in runs. The routine is to run Standard, note the number, run Maximum on the same file, and keep the smaller copy; the result panel holds both sizes.
| Mode | Compression level | Extra step | When to pick it |
|---|---|---|---|
| Standard | 6 | None | First attempt, since it adds no linearization overhead |
| Maximum | 9 | Linearization | A file you also want to open quickly on a slow connection |
Which kind of resume do you have
There is a ten second test that needs no extra tool. Open the PDF in any reader and drag the text cursor across a line of body text. If the line highlights and pastes as words, the page is text. If the cursor snaps to a rectangle around the whole page, or selects nothing, the page is an image.
The second test is zoom. Text stays sharp at any magnification because it is drawn from font outlines, while a scan turns into visible blocks and soft edges. A resume that was printed, signed and photographed fails both tests, and so does one exported from a design tool with the text flattened.
If you would rather look at a file than test a cursor, convert the first page to an image and inspect it at full size. The guide on converting a PDF to JPG without uploading walks through that check.
That answer sets the expectation. Text based resumes respond to a structural rewrite. Scanned resumes do not, because the bytes are pixels and the rewrite cannot touch pixels, so the routes are a fresh export or a smaller scan.
How to shrink a resume PDF in three steps
- Add the file. Drop the PDF into the compressor or click the drop zone. It is read from disk into the tab. A file above 50 MB is refused immediately, before any processing starts.
- Pick Standard or Maximum. Start with Standard. Maximum adds linearization, which helps a reader open the file quickly but does not guarantee a smaller result.
- Download and compare. The result panel shows the size before and after. If the output is not smaller, run the other mode on the same file and keep the smaller copy. The original file is never modified.
There is no quality slider, because no quality is being traded away, and no field where you type a target size in megabytes. The tool cannot be told to hit a number, and the article on why a PDF compressor cannot hit 500 KB explains what a target size would cost.
If the resume is still too large
A rewrite that leaves images alone runs out of room on a scan. These are the steps that work when it does, in the order worth trying.
- Re-export from the source document. A PDF saved many times can carry old objects from every save. A fresh export from Word, Google Docs, Pages or LaTeX writes a clean document that compresses from a better starting point.
- Shrink the photo before it goes in. A headshot that prints at one inch does not need a camera file several thousand pixels wide. Resize the image to roughly the space it occupies on the page, insert it, and export again. This is the single change that moves a photo heavy resume the most.
- Move the appendix out. Portfolios, publication lists and certificates are often the heaviest pages. Use Split PDF with a page range to keep the resume pages, then send the appendix separately when the form accepts a second file.
- Send a link when the form allows one. Some systems accept a URL to a hosted PDF, which removes the size question entirely and is the right answer when the document genuinely needs its images.
None of those steps happen inside the compressor. They are edits to the source document, and the tool is honest about the boundary: it reorganizes a PDF, it does not redesign one.
Split PDF
Keep the resume pages and send the portfolio appendix as a separate file. Enter a page range and get one clean PDF back.
Privacy: a resume is identity data
A resume carries a full name, a phone number, an email address, a city, an employment history and often a photograph. Together that is enough to identify a person and build a profile from, and an unfamiliar converter receives all of it with the file.
Upload based converters keep the file for as long as their policy allows, and policies differ: some delete within hours, some keep files to improve their models, some keep them indefinitely. The retention terms are usually a link away from the upload button.
This compressor takes the other route. qpdf runs as WebAssembly inside the tab, the PDF is read from local storage into page memory, and the output is written back as a download. No account and no email address are involved. The guide on compressing a PDF without uploading it covers how to verify that instead of trusting it: open the developer tools, switch to the network panel, and watch the requests while a document is processed. The engine is about 1.3 MB of WebAssembly, which is code, not your resume.
The trade-off is speed: a large PDF is slower here than on a server, and the real ceiling is the memory of the device. Closing heavy tabs helps when a long document struggles.
What this tool will not do
Each limit below is a case where a different approach is the honest answer.
- There is no target size field. You cannot ask for a specific number of megabytes, and no mode accepts one.
- It does not re-encode images. A scanned resume and a page dominated by a photograph keep their pixels, so their size barely changes.
- It has no OCR. Text inside a scan stays unsearchable, and this tool does not add a text layer that an applicant tracking system could read.
- It does not delete or edit pages. Split PDF extracts a range into a new file; neither tool rewrites the content of your resume.
- It does not add a password or a watermark. An encrypted PDF is refused, so unlock a copy first and compress that copy instead.
- The 50 MB cap is hard. Larger files are rejected in the page, and the workaround is to split the document before compressing the parts.
- It cannot promise the form will accept the result. Only the employer's system decides, and a file that fails for a non size reason will keep failing.