You scan a document, press Ctrl+F to find a word, and nothing turns up, even though the word is plainly on the page. A scan is a picture of text, not text itself. To become searchable or copyable, it needs optical character recognition, or OCR, which reads the letters and adds an invisible text layer beneath the image. This guide explains how to check whether a PDF has that layer, how to turn OCR on when scanning, which settings help it read accurately, and which built-in tools recover text from scans you already have.
What you might be seeing
- Searching a scanned PDF finds nothing
- Dragging over text selects nothing or the whole page
- Copied text comes out as nonsense characters
- Desktop search can't find words inside your scans
Common causes
- No OCR applied
- Most scan tools save image-only PDFs unless you choose a searchable PDF option.
- Poor image for OCR
- Low resolution, skewed pages, faint print and black-and-white settings that break up letters all cut accuracy.
- Wrong OCR language
- OCR set to the wrong language misreads accented letters and special characters.
- Handwriting or complex layouts
- Standard OCR does best on printed text and struggles with handwriting, decorative fonts and dense tables.
How to fix it, step by step
Work through these in order and test a print after each one. Most people are done within the first few steps.
Check whether the PDF has a text layer
Open the PDF and try to select a single word with the text cursor. If nothing highlights, or the whole page highlights as one block, the file is image-only. Press Ctrl+F on Windows or Command+F on a Mac and search for a word you can see. No result confirms OCR hasn't been applied. If text does select but searches still miss words, OCR ran but misread the letters.
Choose searchable PDF when scanning
In the manufacturer's scan app, look for a file type such as searchable PDF or PDF with text, or an OCR checkbox, and set the language to match the document. Some printers offer searchable PDF in the scan file type list on the panel too. Windows Scan, Windows Fax and Scan and Image Capture don't add OCR themselves, so for routine searchable scans the manufacturer's app or printer panel is the easiest route.
Rescan with OCR-friendly settings
OCR can only be as good as the image it reads. Scan at 300 dpi for normal body text, or 400 to 600 dpi for very small print. Use grayscale for faint or colored text; black and white suits crisp, dark text. Turn on automatic straightening if available, keep pages flat against the glass, and clean the glass so specks aren't read as punctuation.
TipZoom to 100 percent on a test scan; if letters look broken to you, OCR will struggle too.
Use built-in text recognition for short passages
For a page or two, your operating system can read the text for you. In Windows 11, open Snipping Tool, capture the area, and select Text actions to copy the recognized text. On a Mac running a recent macOS version, open the scan in Preview and select text directly in the image with Live Text. Recent iPhones and iPads can do the same in the Photos app.
Add OCR to PDFs you already have
Many PDF editors include a recognize text, OCR or make searchable command that processes a whole document. Open the scanned PDF in the editor you already use, run that command with the correct language, and save the result as a new file so the original stays untouched. Then search for a few words you know are present, and check names and numbers, where OCR mistakes matter most.
Let Windows search inside your PDFs
Searchable PDFs still won't show up in File Explorer searches by content unless Windows indexes them. Open Control Panel › Indexing Options, click Modify and make sure your scans folder is included. Then click Advanced, open the File Types tab, select
pdf, choose Index Properties and File Contents and click OK. Indexing may take a while to catch up on a large folder.Work within OCR's limits
OCR handles typed and printed text well, but handwriting, faint stamps, complex tables and multi-column layouts often come out jumbled. For important records, treat the scanned image as the true copy and the text layer as a search aid. Proofread anything you copy out for reuse, especially figures, dates and reference numbers, where a single misread character changes the meaning.
How to keep it from happening again
- Make searchable PDF the default for document scans
- Scan text at 300 dpi with straightening turned on
- Set the OCR language to match your documents
- Spot-check new scans with a quick word search
Common questions
Why does copied text come out as gibberish?
Either the OCR read the page poorly, or the text layer uses an encoding your viewer can't interpret. For scans, rescan at 300 dpi with the correct language and run OCR again. If a PDF that was never scanned does this, the problem lies in how the original file was made rather than in scanning.
Is OCR accurate enough to rely on?
On clean printed pages it's usually very good, but not perfect. Characters such as 0 and O, 1 and l, or 5 and S are common mix-ups. Use OCR to find and reuse text, and check key details against the page image before relying on them in anything important.
Can OCR read handwriting?
Neat printed handwriting is sometimes read reasonably well, especially by the text recognition built into recent phones and computers, but joined-up writing usually isn't. For handwritten pages you need to find later, put descriptive keywords in the file name so a normal file search still locates them.
This guide is general information written for common home and office printers. Menus and button names vary between models, so check your printer’s manual where a step differs. Printer Problems Solved is independent and not affiliated with any printer manufacturer. See our disclaimer.