You open the tenancy agreement the landlord emailed you, press Ctrl+F, type "deposit", and get nothing. Not one match. You try to highlight a sentence to paste into an email, and the cursor just drags a blue box across the page. The words are right there on the screen, and the file has no idea they exist.

That's what a scanned PDF is: a photograph of a page, not a document. Your reader can show it, but it can't search it, copy it, or read it aloud.

The short answer

Here is how to make a PDF searchable: run OCR on it. OCR, short for optical character recognition, reads the words in the picture of each page and writes an invisible layer of real text behind it. The page looks exactly the same afterwards, but Ctrl+F works, you can select and copy a paragraph, and a screen reader can read it. You can make a PDF searchable free in your browser with Xplosole's PDF Make Searchable tool. No upload, no signup.

A document on a flatbed scanner with the lid open

Photo by Olena Kholina on Unsplash: a document on a flatbed scanner.

First, check whether your PDF is actually a scan

Not every stubborn PDF is a scan. Some are real text files with a broken font, and OCR won't help those. The test takes ten seconds: open the file, try to drag your cursor across a line of text, then press Ctrl+F (Cmd+F on a Mac) for a word you can see on the page.

If the words highlight, your PDF already has text, and the problem is somewhere else. If the cursor drags an empty box and the search finds nothing, the page is a picture. That is the file this guide fixes.

Make it searchable in five steps

The tool runs entirely in your browser: the OCR runs on your device with Tesseract, the open-source engine, and only the language data downloads the first time you use a language.

  1. Drop the scanned PDF into the upload area. Say a 17-page tenancy agreement, a 3.8 MB scan from the office copier.
  2. Pick the document's language. English, Urdu, Arabic, Hindi, Spanish, French, German, Portuguese, Turkish and Simplified Chinese are all there. If the document mixes two, pick both, and the tool reads each line in the right one.
  3. Leave "Skip pages that already have text" ticked. Most scanned files hide a stray typed page in the middle, a cover sheet or a printed form, and reading those again would put every word in twice, so your searches and copied text would come out doubled.
  4. Click "Make searchable". Progress shows page by page, and you can cancel. Each page is rendered upright at about 250 DPI and read by Tesseract, which writes the invisible text layer in a glyphless font laid exactly over the page, the same approach OCRmyPDF uses. It works for every script, including Arabic and Chinese.
  5. Download the result. It looks the same as the scan you put in, but try Ctrl+F on it now. The word is there.

If the scan came from a shared office machine, it may carry metadata about the scanner and who ran it, so remove metadata from a PDF before you send it on. And if the scan includes blank pages, delete pages from a PDF first, so the job runs faster on the pages that matter.

When the words come out wrong

OCR is only as good as the picture it reads. Clean scans at 200 to 300 DPI of printed text usually read very accurately: 200 DPI is the floor, 300 is the sweet spot, and going higher mostly makes the file bigger without making the reading better.

Three things break it. Blurry photos read badly, because the letters smear into each other. Low-resolution scans do the same. And handwriting reads poorly, full stop, no matter how neat it is. If your scan is a phone photo taken in a dim room, rescan it flat on the scanner before you blame the tool. Rotated or cropped pages are fine, the tool straightens those itself.

One more honest limit: a printed letter with a handwritten note in the margin reads cleanly except for the handwriting, which comes out as guesswork or nothing at all. No setting fixes that. OCR the printed pages and type the handwritten lines yourself.

And remember: making a PDF searchable doesn't make it editable. The visible words are still a picture, and changing the text itself is a different tool and a different afternoon.

The other ways, ranked

Search for OCR PDF online free and most of what ranks uploads your file to a server first. The browser tool above is the best option for most people: it hands you back the same PDF, searchable, without an account and without uploading a contract to someone else's server. Two of the other routes win in specific cases.

Google Drive is the best free route when you only need the words out. Upload the PDF, right-click, choose Open with, then Google Docs. Drive runs OCR and opens the recognised text in a new document. The catch: you get editable text, not a searchable PDF. The page images and layout are gone. Use it to quote a passage or pull text into your notes, not to archive the original.

Adobe's free online OCR works, with strings attached. You upload the file, the OCR runs, and you sign in to download the result. One document at a time, and the file goes to Adobe's servers. Fine for one letter, irritating for a folder of them.

On a phone, start with a scan app. If the paper is still on your desk, Adobe Scan photographs the page and runs text recognition on it. That's the right tool for a receipt you're holding, not for a PDF already on your computer.

Acrobat Pro is the paid route, and it is only worth it if you do this every week. Its Recognize Text command does the same job well. For a one-off contract, paying a monthly subscription to run it once is a bad trade. For an office that OCRs archives daily, it's the correct tool.

So the ranking is simple: a scanned PDF you already have means the free browser tool; words you need out of a scan right now means Google Drive; a stack of scans every week means Acrobat Pro.

Questions people actually ask

Will my PDF look different after OCR?

No. The visible page is untouched: the text layer is invisible, laid exactly over the scan, so the file looks identical and only behaves differently. Search, select, copy and screen readers all start working.

Is it really free, and does my file get uploaded?

Yes to free, no to uploading. The tool runs Tesseract in your browser, so the file never leaves your device. Only the language data downloads, the first time you use a language, and your browser caches it after that. There's no account, and no daily limit is stated on the tool page, because there's nothing server-side to limit.

Can I make scanned PDF searchable on my phone?

You can, but a phone is the wrong place for a long document. The tool runs in a mobile browser, so a short scan works fine. For a 40-page file, use a computer: the language data downloads faster, and you're not fighting a small screen when a page reads badly. If the paper is still in your hand, scan it with an app like Adobe Scan first, then OCR the PDF on a computer.

What if my document is in two languages?

Pick both before you run it. A document that mixes two, a Spanish contract with English terms, reads each line in the right one. Picking the wrong language is the most common reason a clean scan reads badly.

Do the ten-second test on your scans now, not on the day a form is due. A PDF that can't be searched will wait quietly in your inbox for months, and it always announces itself at the worst moment: a deadline, a clause you need to quote, a form that wants the exact wording. Run OCR while you have time, and the file stops being a photograph and starts being a document.