How to Check If a PDF Is Valid or Corrupted Online for Free

To check if a PDF is valid or corrupted online for free, open the file in the Validate PDF tool and run the check: it reads the file's structure, its cross-reference table, its page tree and its fonts, then reports whether the document is well-formed or where it breaks. A file that passes is structurally sound and readable; a file that fails has damage you can usually name — a truncated download, a broken reference table, a missing font — and the fix depends on which check failed. The check works in your browser, so the file is read on your device and you get the result in seconds, before you send the document on or try to repair it.

What "Valid" Actually Means for a PDF

A PDF is not a single block of content. It is a small filing system: a header that names the format version, a body made of numbered objects (pages, fonts, images, text streams), a cross-reference table that records where every object sits in the file, and a trailer that points the reader to that table and to the document's root. A reader opens a PDF by following those pointers, not by reading the file from top to bottom like a book. That is why a PDF can look fine in one viewer and fail in another, and why a file that opens can still be damaged underneath.

Validation means checking that this filing system is intact. It does not judge whether the document is well written, well designed or legally correct. A one-page invoice with a typo passes validation happily; a beautifully designed 200-page report with a broken reference table fails it. Keeping those two ideas apart saves a lot of confusion: validation answers "is this file structurally sound?", not "is this a good document?"

There is also a difference between "valid" and "standard-compliant". A file can be readable by every common reader yet bend a rule in the PDF specification, and a file can follow the specification strictly yet use a feature an old viewer does not support. For everyday purposes — sending a contract, submitting a form, archiving a report — the practical question is the simpler one: does the file open reliably, show every page, and print the way it looks on screen? That is the question a validator answers.

What a Validator Checks

A validator works through the file in roughly the order a reader does, and each check catches a different kind of damage. Knowing what each check does matters, because the check that fails tells you what kind of repair — or what kind of replacement — you need.

The cross-reference table deserves special attention, because it is the check that fails most often on real-world files. It is essentially an index at the end of the file, and it is also the last thing written when a PDF is created. Anything that interrupts the write — a crashed export, a killed upload, a download that stops early — damages the index first, while the pages themselves may be perfectly intact. That is why a viewer can sometimes "repair" a file on open: it ignores the broken index and rebuilds it by scanning the objects directly. A rebuilt index is a rescue, not a clean bill of health, and the file should be re-saved properly before it is relied on.

The bar lengths above show check order only — step 1 runs first, step 5 last — not a measurement of anything. A failure at an early step usually stops the later checks from meaning much: if the header is unreadable, the page tree cannot be trusted either, so fix the earliest reported problem first and re-check.

Signs a PDF Is Corrupted

Most people reach a validator because something already looked wrong. The symptom often points straight at the damage, which is useful when you have to explain the problem to whoever sent the file.

One symptom deserves a warning of its own: a file that opens fine on your computer but fails elsewhere. That usually means your viewer is quietly repairing damage on open while a stricter viewer — a government portal, a print shop, an archival system — refuses it. "It opens for me" is not proof a file is sound. Validation is the neutral test that settles it, and it is worth running before a deadline rather than after a rejection.

Checking a PDF, Step by Step

1. Open the Validate PDF tool. Go to the Validate PDF page in your browser.
2. Choose the file. Pick the PDF from your device. If you suspect a download problem, download a fresh copy first and check that copy, so you are testing the file you will actually send.
3. Run the check and read every reported problem. Do not stop at pass or fail: note which check failed, because that names the damage and the fix.
4. Open the file in a viewer and spot-check pages. Look at the first page, a middle page and the last page, and any page with images or unusual fonts. Validation plus a visual check together catch more than either alone.
5. Act on the result. A pass means the structure is sound. A fail means repair, re-export or a fresh copy from the sender, as the next section explains.

When Validation Fails: What to Do

Start with the cheapest fix. If the file arrived by download or email, get a fresh copy first — interrupted transfers are the most common cause of a failed check, and a second download costs nothing. Compare file sizes: if the new copy is larger than the failed one, the first copy was almost certainly truncated, and the new copy may already pass.

If a fresh copy still fails, and you have the original document (a Word file, a design file, an invoice in an accounting app), re-export it. Re-exporting rebuilds the whole structure — header, objects, page tree, reference table — from scratch, which fixes every class of structural damage at once. When you re-export, embed fonts rather than referencing them, and avoid exotic compression options unless you know the recipient's viewer supports them.

If neither a fresh copy nor a re-export is available, try a repair. A repair tool such as Repair PDF rebuilds the damaged parts of the structure — typically the reference table — from the objects that are still readable, and saves a new file. Repair works best when the pages themselves are intact and only the index is broken. It cannot restore pages whose content is genuinely missing, and it cannot reconstruct a font that was never embedded. After any repair, validate the repaired file again: a repaired file that passes is usable, but keep the damaged original until you have confirmed the repaired copy shows every page correctly.

When the damage is in the content — blank pages, missing images, garbled text that no viewer can decode — no validator or repair tool can invent what is gone. The only honest fix is a new copy from the source. Say so plainly to whoever sent the file, and name the failing check: "the page tree lists 40 pages but only 32 have readable content" gets a useful replacement faster than "the file is broken".

When Validation Passes but the File Still Looks Wrong

A pass rules out structural damage, so the problem lies somewhere else, and the fix changes accordingly. Work through the possibilities in order.

First, suspect fonts. A PDF that references a font without embedding it looks perfect on the machine that created it — where the font exists — and wrong everywhere else, where the viewer substitutes something else. The text is still there, which is why validation passes, but the lettering changes, spacing shifts, and special characters can vanish. The fix is on the creation side: re-export with fonts embedded, or convert the text to outlines if the document is meant for print.

Second, suspect the viewer. Old or minimal PDF readers sometimes render transparency, layered graphics or newer compression poorly, producing blank patches or misplaced elements in a structurally sound file. Open the same file in a second, current viewer — a modern browser's built-in viewer is a good neutral test — and if it renders correctly there, the file is fine and the first viewer is the problem.

Third, suspect the content itself. A page can be structurally perfect and still contain a low-resolution image, a scanned page photographed at an angle, or text laid out badly. Validation never judges content quality; a visual check does. If the file passes and renders identically in two viewers, what you are seeing is what the document actually contains, and the remedy is a better source document, not a repair.

Finally, check for optional content the validator does not weigh: attachments, form fields, digital signatures and annotations all sit outside the core page structure. A form whose fields do not work, or a signature that shows as invalid, is a different problem from corruption, and it needs the form or signing tool that created it, not a structural repair.

The Scanned and Image-Only PDF Caveat

Scanned PDFs need a separate mental model. A scanned page is a photograph of a page: it has no text layer, no selectable words and no embedded fonts, because there is no text in it at all — only pixels. A validator will usually pass a scanned PDF, because its structure is simple and intact, but that pass says nothing about whether the scan is readable, straight, or complete. A blurry, skewed or half-cut scan passes validation and is still useless.

The reverse confusion is just as common. Someone opens a scanned PDF, finds they cannot select or search its text, and concludes the file is corrupted. It is not: it was never a text document. The remedy is optical character recognition (OCR), which adds a text layer behind the image, or a fresh scan at a sensible resolution. Until OCR runs, searching, copying and screen-reader access will not work, and no amount of validation or repair changes that, because there is nothing structurally wrong to fix.

When you validate a scanned file, judge it by two separate tests: does it pass validation (structure sound?), and can you read every page at normal zoom (scan usable?). Both must pass before the file is fit to send, and a failure of the second is a scanning problem, not a corruption problem. If you need to read the contents before deciding, open the file in a viewer such as PDF Viewer and page through it visually.

Common Mistakes

1. Trusting "it opens for me". Viewers silently repair damaged files. A file that opens on your machine can still fail a portal, a print shop or an archive. Validate before you send, not after a rejection.

2. Repairing before re-downloading. Repair is a rescue for a file with no good copy. If a fresh download exists, it is faster, cleaner and safer than any repair.

3. Deleting the original after a repair. Keep the damaged file until the repaired copy has been validated and visually checked page by page. A repair that drops a page silently is worse than the original damage.

4. Treating a pass as a quality guarantee. Validation checks structure, not content. A passing file can still have blurry scans, substituted fonts or bad layout — check it visually too.

5. Calling a scanned PDF corrupted because its text cannot be selected. Image-only pages have no text layer by nature. That is a job for OCR or a re-scan, not for repair.

6. Sending the repaired file without re-validating it. The repaired copy is a new file. Validate it, then check its pages, then send it.

Frequently Asked Questions

For the surrounding document workflow, pair validation with Repair PDF when a file fails, and PDF Viewer to page through a file visually. More step-by-step tool guides live on Asad Webs blog.