Google Drive has had a genuinely free OCR path for years, and most people who store scans in Drive have never noticed it: Open with → Google Docs. For a short, clean scan it's often all you need. For anything longer or messier it fails in predictable ways, so this guide covers both the steps and the exit ramp.
The three steps
- Upload the scanned PDF to Drive (drag it in, or New → File upload). Images work too - a photographed page is fine.
- Right-click the file → Open with → Google Docs. This is the OCR step: Drive converts the file to a Doc, recognising text on image-only pages as it goes.
- Review the result. The Doc typically shows each page image with the recognised text beneath it. Proofread anything that matters - recognition quality tracks scan quality, and numbers and names are where errors hide.
From there you can copy the text, keep the Doc, or paste the text wherever it was headed.
Where Drive's OCR breaks down
Drive's conversion is tuned for short, simple documents, and its failure modes are consistent:
- Long scans. A multi-hundred-page scan is a lottery - conversions can come back partial or fail outright. Drive's OCR is happiest under a few dozen pages.
- Layout. Tables, multi-column layouts, and forms come back as interleaved text in roughly reading order. The words survive; the structure doesn't.
- Low-quality pages. Skewed photocopies, faxes, and low-resolution photos produce garbled runs. There's no way to tune the recognition.
- Handwriting. Print does fine; handwriting is hit and miss. See can AI read handwriting in PDFs for what's realistic.
If your scan hits one of these walls, a dedicated OCR tool (OCRmyPDF, Acrobat) will do better - or skip the conversion entirely, below.
When the goal is answers, not a converted file
Most Drive OCR journeys start with a question - "what does this scanned contract say about notice?" - and the conversion is just a means to it. You can skip the middle step. Sidenote is a browser extension that reads the PDF you have open, including PDFs opened from Drive, and runs OCR automatically when pages are image-only (a paid feature, included in the 7-day trial). Ask the question directly and every claim in the answer carries a citation that scrolls to the exact passage, so you can check the scan itself rather than trusting recognition you haven't proofread.
That verification step matters more for scans than for born-digital files: OCR errors are silent, and a misread "£15,000" is exactly the kind of thing you want a click-to-source citation to catch. For the broader Drive workflow - summaries, multiple files, mixed formats - see the best AI tools to chat with Google Drive.
Drive OCR vs the alternatives
| Google Drive | OCRmyPDF / Acrobat | Sidenote | |
|---|---|---|---|
| Cost | Free | Free / paid | Paid feature, in trial |
| Output | New Google Doc | Searchable PDF | Cited answers in place |
| Long scans | Unreliable | Handles well | Handles well |
| Layout preserved | Rarely | Yes (image kept) | Yes (reads in place) |
| Best for | Short, simple scans | Archiving searchable files | Reading and questioning scans |
Rule of thumb: Drive for a quick one-off, a dedicated OCR tool when the file itself needs fixing, and an OCR-aware reader when what you actually want is the answer.