You scanned a stack of paperwork, opened the PDF, and tried to search it — nothing. You can’t select the text, you can’t copy a sentence, and Ctrl+F finds zero matches. That’s not a broken file: a scan is a photo of a page, so the words are pixels, not text. The fix is OCR — it takes about a minute — and this guide walks you through it: what OCR really does, how to get the most accurate result, how to do it on any device, what it can’t read, and everything a searchable PDF then unlocks.
Why a scan isn’t searchable
When a scanner or phone camera captures a page, it saves an image — millions of dots that look like letters to you, but mean nothing to the computer as text. There’s no text layer to search, select or copy; a PDF made this way is really just a picture wrapped in a PDF. The quickest way to check your own file: open it in any reader and try to select a line with your cursor. If you can’t highlight the words, it’s a scan and needs OCR.
What OCR actually does
OCR stands for Optical Character Recognition. It examines the image, recognises the shapes as letters and words, and writes an invisible layer of real text behind the picture. The page looks exactly the same — nothing about the appearance changes — but the words underneath are now selectable and searchable. You simply gain a text layer you couldn’t see was missing. Good OCR also keeps the layout, so a searchable scan of a form or letter still reads and prints exactly as before.
How to make a scanned PDF searchable
Ream’s OCR PDF adds the text layer in three steps:
- Add your scanned PDF to OCR PDF.
- Choose the language — English, German, or both — so recognition is accurate.
- Run it and download. The new PDF looks identical, but Ctrl+F, select and copy all work now.
The result is an ordinary PDF that happens to be searchable, so it opens in every reader on every device — Windows, Mac, iPhone and Android — with nothing to install. Just scanned the pages? Scan to PDF captures them with your camera first; then run OCR on the result.
Get the most accurate result
Recognition quality depends almost entirely on the input. A few habits make a big difference:
| Do this | Why it helps |
|---|---|
| Scan around 300 dpi, in good light | Sharper shapes are recognised far more reliably than blurry, low-res scans. |
| Match the language | Reading English text with a German dictionary (or vice-versa) produces odd results — pick the right one, or both. |
| Let it straighten (deskew) | Skewed pages hurt accuracy; Ream deskews as it works, so crooked scans come out straight. |
| Keep the page flat and uncropped | Shadows and cut-off edges from phone photos confuse the recogniser. |
| Compress afterwards, not before | Compress the sharp original after OCR so recognition works on full detail — see compressing a PDF. |
How to OCR a scan on any device
A dedicated tool is the most reliable way to get a genuinely searchable PDF, but your devices have built-in tricks worth knowing — with one honest catch: several of them let you select text on screen without actually saving a searchable file.
- Any device (browser): OCR PDF adds a real, saved text layer to the PDF itself — the same result everywhere, nothing to install.
- Google Drive (free): upload the PDF or image, right-click → Open with → Google Docs. Drive OCRs the text into a new Doc. Great for pulling the words out — but it reflows into a document, it doesn’t hand back a searchable copy of your original PDF.
- Mac: macOS Live Text lets you select and copy text straight from a scanned page in Preview — handy for a quick copy, but it doesn’t save a searchable PDF layer on its own.
- iPhone & Android: iOS Live Text and Google Lens recognise text in a photo or scan so you can copy it; for a searchable file, run the PDF through OCR PDF in your mobile browser.
- Windows: there’s no built-in PDF OCR, though OneNote can “Copy text from picture”. For a searchable PDF, a dedicated tool is the simplest path.
The distinction that trips people up: “I can select the text on my Mac/iPhone” isn’t the same as “the PDF is searchable everywhere.” Live Text reads the image live, on that device. To get a file that’s searchable for anyone, in any reader, you need OCR written into the PDF — which is exactly what OCR PDF does.
What OCR can — and can’t — read
Modern OCR is remarkably good on printed text, but it isn’t magic. Set expectations:
- Printed text: excellent, especially at 300 dpi in a supported language.
- Handwriting: unreliable. Neat block capitals sometimes work; cursive rarely does.
- Tiny, faded or decorative fonts: hit-and-miss — expect a few misreads.
- Complex layouts (multi-column, tables): the text is recognised, but reading order can jump around.
- Numbers matter: always spot-check figures after OCR — a misread digit in an invoice or statement is worse than an obvious typo.
OCR, PDF to Word, or re-typing?
These solve different problems. OCR keeps the page looking exactly as scanned and just makes it searchable — ideal for records, contracts and archives you want to keep as-is. PDF to Word goes further and rebuilds an editable document — run OCR on the scan first, then convert; our guide on what survives a PDF-to-Word conversion explains what to expect. Re-typing is only worth it for a paragraph or two — for anything longer, OCR plus a quick proofread wins every time.
What a searchable PDF unlocks
Adding a text layer unlocks the rest of your toolkit. You can copy quotes straight out of the document, find any figure or name in seconds, convert it to an editable Word file, or pull the raw words out with PDF to Text. And because scans are bulky, it’s worth following OCR with Compress or Grayscale — a searchable, right-sized archive of your paperwork is far more useful than a drawer full of images. It also makes documents accessible: screen readers can finally read a scanned page aloud once it has a real text layer.
Is it private, and is it free?
Recognising text needs a real OCR engine, so OCR PDF runs on Ream’s server (marked with a “Server” badge) rather than in your browser — but your file is processed in an isolated, network-cut sandbox and deleted immediately afterwards, never stored or shared. It’s free, with no account and no watermark. If you’re weighing up any online tool for sensitive paperwork, our guide on whether online PDF tools are safe explains what to check.
Frequently asked questions
How do I make a scanned PDF searchable? Run it through OCR PDF: add the file, pick the language, and download — the new PDF looks identical but is fully searchable and selectable.
Will OCR change how my document looks? No. The text layer is invisible and sits behind the image, so the page looks identical — it just becomes searchable, and any skew is straightened.
How accurate is OCR? On a sharp, straight scan of printed text in a supported language, very accurate. Blurry, skewed or low-resolution scans produce misreads, so always spot-check numbers.
Can OCR read handwriting? Usually not reliably. Neat block capitals occasionally work; cursive rarely does. OCR is built for printed text.
Which languages are supported? English and German, individually or together. Matching the language to the document improves accuracy.
How do I OCR a PDF in Google Drive? Upload it, right-click → Open with → Google Docs; Drive extracts the text into a Doc. It’s free, but you get an editable Doc, not a searchable copy of your original PDF — use OCR PDF for that.
What if my PDF already has some text? Pages that already contain real text are left as they are, so nothing searchable is changed.
Is it free and private? Yes — free, no sign-up, no watermark, and your file is processed in isolation and deleted right after.
One quick pass through OCR turns a dead scan into a document you can search, copy and reuse. Do it once, right after scanning, feed it a sharp and straight image, and everything you do with that file afterwards — converting, compressing, archiving — gets easier.
