Inspect a PDF
See whether each page has real text or is a scanned picture, which fonts are embedded, and what a reader objects to. Nothing is uploaded.
Processed on your device — no upload
How to use it
- Choose the PDF you want to know about.
- Press inspect. Every page is read, which takes a moment on a long document.
- Read the verdict per page, the embedded fonts, and what produced the file.
How it works
A PDF makes no distinction between a page of typed text and a photograph of a page. Both are drawing instructions, both print identically, and on screen they are indistinguishable until you try to select a word and find there is nothing to select.
This reads the text layer of every page and counts what is there. A page of a typed report returns a few hundred characters. A scanned page returns none at all — not because the words are hidden, but because no words were ever stored: the page is one picture, and the letters in it are pixels.
A few characters is not text. Scanners stamp page numbers, and software adds headers, so a scanned page often carries a handful of characters without being searchable in any useful sense. Pages are called text only past a small threshold, and the exact count is shown beside each one so you can disagree with the judgement.
The font list comes from the same read. Fonts are named in the file — Times-Roman, Helvetica, an embedded subset with a tag in front of it — and a page that draws no text names none, which is a second confirmation of the same finding from a different direction.
Nothing is modified and nothing is written. That is what lets this open files the other tools here would refuse: when a document is damaged, the error is the answer rather than a failure, so the reader's own words are shown rather than replaced with something friendlier that says less.
Making a scan searchable needs OCR, which this site does not do — a browser reads a difficult scan far less accurately than a cloud service, and a bad transcription is worse than an honest picture. If the answer here is “picture only”, that is the work you need next, elsewhere.
A worked example
A contract comes back from the other side and you need to check one clause before signing. Ctrl+F finds nothing — not the clause, not any word in the document, not even a word you can see on the screen in front of you.
Inspecting it says: twelve pages, none with text, no embedded fonts, produced by a scanner. Somebody printed the contract, signed it, scanned it, and sent back a stack of photographs. Nothing is wrong with the file and nothing you do to it will make Ctrl+F work, because there are no words in it to find.
That takes a few seconds and settles the question. The alternative — zooming in, trying to drag over a line, doing it again on a different page in case that page was different — takes ten minutes and only ever tells you about the page you tried.
The mixed case is the one worth catching. A twenty-page report where pages 1 to 18 are typed and 19 and 20 were scanned and inserted looks entirely normal and searches entirely normally, right up to the point where it silently fails to find anything on the last two.
Questions
How do I know if a PDF is scanned or has real text?
Open it here and read the per-page verdict. A page with real text returns hundreds of characters; a scanned page returns none, because it is a single image and the letters in it are pixels rather than characters. The old advice — zoom in and try to select a word — works on one page at a time and tells you nothing about the other two hundred.
Why does my reader say "not a valid PDF document"?
Usually one of three things: the file was truncated in transit and its cross-reference table points past the end, it is not a PDF at all but something renamed, or it is an HTML error page that was saved with a .pdf extension. Inspecting it shows which — a file that opens here and reports its pages is structurally fine whatever the other reader said, and a corrupt one fails with the specific reason rather than a shrug.
What properties of the file does this report?
Page count, the size of each page in points, how many characters of text each page holds, the embedded fonts, the producer that wrote the file, its PDF version, and whether it is encrypted. It will analyse a document it cannot otherwise do anything with, because reading is all it does — nothing here rewrites the file.
How can I see which fonts a PDF uses?
They are listed after inspecting. Names may carry a six-letter prefix and a plus sign, like ABCDEF+Arial, which marks an embedded subset — only the glyphs the document actually uses were included, which is why the file is smaller and why the font cannot be extracted and reused.
A page has a few characters but I cannot search it. Why?
Those characters are almost certainly not the page content. Scanning software stamps page numbers, and some scanners add a header line. The page itself is still a picture. That is why a small character count is reported as a picture rather than as text, and why the count is shown rather than only the verdict.
Can this make a scanned PDF searchable?
No, and that is deliberate. Turning a picture of a page into text is OCR, and OCR in a browser reads a difficult scan far less accurately than a cloud service does. A transcription with errors scattered through it is worse than no transcription, because you cannot see which words are wrong.
Does it work on a password-protected PDF?
It reports that the file is encrypted, which is often the answer somebody needs. Reading the pages inside a document that is genuinely locked requires the password, and this tool does not ask for one.
Why does the page count differ from what my reader shows?
It should not. If it does, the file most likely has a damaged cross-reference table and the two readers have recovered different amounts of it. That is worth knowing before you send the document anywhere.
Is my document uploaded to inspect it?
No. The file is read in this tab by your own browser and never leaves your device. The site is static files with no backend, so there is no endpoint to send it to — which matters here more than most places, since the documents people need to inspect are usually the ones that arrived from somewhere else.
Is Inspect a PDF free to use online?
Yes — free, with no account, no daily limit and no watermark. It runs online in your browser, and because the work happens on your own device it keeps working offline once the page has loaded.
Related PDF tools
Extract Text from PDF
Pull the words out as plain text
open$termiva compress document.pdf --balancedCompress PDF
Smaller file, text still text
open$termiva fit report.pdf --under 2mbCompress PDF to a Size
Name the limit, not the quality
open$termiva scrub document.pdf --allRemove PDF Metadata
Erase hidden author and app data
open