Runs in your browser with pdf.js and pdf-lib
PDF Analyzer — check a PDF before you print, share or send it
One upload gives a full report: what makes the file big, fonts that aren't embedded, images too blurry to print, passwords and restrictions, and hidden extras like comments, attachments, scripts and earlier versions. Every finding comes with the fix.
- PDF files aren't uploaded to our server
- Where the file size goes
- Print checks: fonts and image dpi
- Hidden content found
Analyze a PDF
Add a PDF and get a full report in seconds: what is inside, what could cause trouble, and what to fix.
Drop your PDF here
or browse: one PDF, any number of pages
Your pages
Each page with its size, words and images. Pages with problems are marked; click a page to see it full size.
No page has a finding. Untick Only pages with a finding to see every page.
What this tool does
A PDF looks fine on screen. This shows what's underneath.
Missing fonts, blurry images, blocked printing and hidden comments don't show until someone else opens, prints or reads the file.
What you get
- A short list of findings, the important ones first, each with what it means and how to fix it.
- Where the file size goes: images, fonts, page content, attachments and metadata.
- Print checks: fonts that aren't embedded and the real dpi of every image as placed on the page.
- Before-you-share checks: author details, comments, attachments, scripts, hidden layers and earlier saved versions.
- Text facts: words, reading time, scripts such as Gujarati or Devanagari, most used words and readability.
- The whole report as an HTML file you can print, plus JSON, CSV and TXT.
What it can't do
- Scan for viruses. It reports JavaScript and attachments, but doesn't judge whether they are harmful.
- Read scanned pages. It tells you which pages are scans; PDF OCR adds the text.
- Certify PDF/A or PDF/UA. It shows what the file claims, not a full standards check.
- Open a PDF that needs a password to open. Remove it with Unlock PDF first.
How to analyze a PDF in three steps
Add the file, read the findings, fix or save.
Add your PDF
Drop it on the upload area or click Select PDF. The report starts by itself; a 100-page file takes a few seconds.
Read what we found
- Red items need a look before you send or print.
- Blue items are worth knowing.
- Green items passed.
Fix it or save the report
Follow the link on a finding to the tool that fixes it, open the detail sections, or download the report.
What to look at for your situation
The part of the report that answers each question.
| Situation | What matters | Where to look |
|---|---|---|
| Sending a file to a print shop | Embedded fonts, image dpi, page size, bleed. | Fonts and Images findings, then Every page for dpi per page. |
| Email says the file is too big | What takes up the space. | Where the file size goes; if images lead, use Compress PDF. |
| Sharing a contract or report outside | Author name, comments, earlier versions. | Privacy findings; fix with PDF Metadata and Flatten PDF. |
| PDF from an unknown sender | Scripts, attachments, links. | Security findings and Links, bookmarks and attachments. |
| Can't print or copy from a PDF | Which actions are blocked. | Security and permissions. |
| Quoting a translation or typing job | Words per page, scanned pages, scripts. | Text section and the pages CSV. |
| Government or college upload rules | Size, PDF/A, page size, no password. | Document details and the findings list. |
The details
How each check is worked out
Measured from the file itself, not guessed.
File size
Every stored object is sorted by what it is: images, font programs, page drawing instructions, attached files or metadata. The rest is the PDF's own structure.
Fonts
Each font on each page is checked for an embedded font program. A name like ABCDEF+Arial means only the letters used are embedded, which is normal.
Image dpi
The page's drawing steps are followed to find how big each image is placed. Pixels divided by inches gives the dpi; under 150 is flagged for print.
Permissions
The eight permission flags of an encrypted PDF are read: printing, full-quality printing, copying, editing, comments, forms, screen readers and page assembly.
Hidden content
Comments, attachments, scripts, layers and form answers are counted. Each extra end-of-file marker shows a later save added on top of the original.
Text
Words are counted the way your browser splits text, so Hindi, Gujarati and Chinese count correctly. Readability uses the Flesch formula and only for English.
When the report surprises you
What a finding usually means, and what to do.
Zero words, but I see text
The pages are pictures of text, from a scanner or a phone.
Run PDF OCR, then analyze the searchable copy.
Images below 150 dpi
Small images were stretched, or the file was saved for screen or email.
Export again from the original with full-size images and "high quality print" settings.
A small file is still big
Large images, full fonts, attachments or earlier saves inside.
Check Where the file size goes, then Compress PDF.
Printing or copying blocked
The author set restrictions with an owner password.
If it's your file, remove them with Unlock PDF.
"Saved on top of itself"
Your PDF app adds each save to the end of the file.
Use Save As, or print to a new PDF, before sharing sensitive files.
The PDF won't open
It needs a password to open, or it is damaged.
Use Unlock PDF or Repair PDF first.
PDFs that cause no trouble later
Habits that keep the findings list short.
Export, don't print to PDF
Save as PDF from Word, Docs or your design app. It keeps text, links and tags that a printed PDF loses.
Embed fonts for print
Turn on "embed fonts" or a "print" or PDF/X preset when the file goes to a printer.
Place images at 300 dpi
An image printed 10 cm wide needs about 1,200 pixels across to look sharp.
Clean before sharing
Remove comments, old versions and author details from documents that leave your organization.
Set a title and language
Fill in the document title and language before exporting, so screen readers and search results show them.
Be careful with scripts
A PDF from someone you don't know that contains JavaScript or attachments deserves a second look.
Common questions about the PDF Analyzer
If something's not covered here, the section above likely has it.
Is my PDF uploaded to a server?
No. The PDF is opened and analyzed in your browser with pdf.js and pdf-lib, and the file isn't uploaded to our server. The report is built on your device, and the downloads are made there too.
What does the analyzer check?
Whether every page has real text, whether fonts are embedded, how sharp the images are for print, what makes the file big, passwords and restrictions, hidden extras such as comments, attachments, JavaScript, hidden layers, filled form fields and earlier saved versions, accessibility basics, links, and text statistics. Each finding says what it means and how to fix it.
What does "font isn't embedded" mean?
The PDF names the font but doesn't carry it. A computer without that font shows the text in a substitute, so line breaks and spacing can change, and print shops usually reject such files. Export the PDF again from the original app with fonts embedded. The 14 standard PDF fonts, such as Helvetica and Times, are built into readers and are listed separately.
How is the image dpi worked out?
For every image on a page, the pixel size is divided by the size it is placed at, in inches. 300 dpi or more is print quality; under 150 dpi usually prints soft. Small images under an inch, such as icons and bullets, are left out of the check.
What does "saved on top of itself" mean?
Many apps save changes by adding them to the end of the PDF instead of rewriting it. The earlier content can still be inside the file and can sometimes be recovered. Before sharing a sensitive document, use Save As in your PDF app or print it to a new PDF to get a clean copy.
Can it analyze password-protected PDFs?
PDFs that only restrict printing or copying open fine, and the report lists exactly which actions are blocked. A PDF that needs a password to open can't be read; remove the password with Unlock PDF first.
Why is the word count zero?
The pages are images, usually a scan, so there is no text to count. Run the file through PDF OCR to add a text layer, then analyze the searchable copy.
Build notes
How this PDF Analyzer works under the hood
I wrote up how to build a browser-based PDF analyzer: reading metadata and permissions with pdf.js, extracting text, and building a report on the user's own device.
How to Build a Browser-Based PDF Analyzer Using JavaScript
Covers metadata inspection, permission checks, text extraction and report generation with client-side JavaScript, keeping documents on the user's device.