How to Batch OCR Multiple PDFs (Free & Paid Ways That Scale)

One scanned PDF is a chore. Forty of them - a due-diligence folder, a box of digitised invoices, years of committee minutes - is a workflow problem, because every tool that handles one file gracefully becomes miserable when you have to click through it forty times. Batch OCR is the fix, and there are three routes that genuinely scale, plus one way to skip the conversion entirely.

Route 1 - OCRmyPDF (free, scriptable, the power option)

OCRmyPDF is a free, open-source command-line tool that adds a text layer to a scanned PDF without touching the page images. Batch is a one-liner over a folder:

for f in scans/*.pdf; do ocrmypdf "$f" "ocr/$(basename "$f")"; done

It deskews, it can run files in parallel, it skips pages that already have text, and the output is a proper searchable PDF suitable for archiving. The cost is the terminal itself - if that's a dealbreaker, keep reading.

Route 2 - Acrobat's Action Wizard (paid, point and click)

Acrobat Pro's Action Wizard lets you build an action - Recognize Text, save - and apply it to an entire folder in one run. If your office already pays for Acrobat this is the lowest-friction route, and the same action can also compress or standardise the files while it's in there. Without an existing licence, the subscription is hard to justify for OCR alone when the free routes exist.

Route 3 - NAPS2 (free, graphical batch queue)

NAPS2 is a free, open-source scanning app for Windows, macOS and Linux with a built-in OCR batch queue - the middle ground when you want clicks rather than a terminal but Acrobat isn't on the table. Point it at your files, pick the language, and let the queue drain.

The reframe: why are you OCRing forty files?

Sometimes the archive genuinely needs fixing - compliance, search across a document management system, future-proofing. Then batch conversion is the right job, use the routes above.

But often the forty files are a means to a handful of questions: which of these contracts auto-renew? what did the committee decide about the roof? which invoices mention the disputed PO? For that, converting every file is the long way round. Sidenote runs OCR automatically on each scanned PDF as you ingest it (a paid feature, included in the 7-day trial), and a collection lets you ask one question across the whole set - with each answer carrying a citation that names the source document and scrolls to the exact passage when you open it.

The practical difference shows up at verification time. A batch-OCR'd archive gives you searchable files you still have to open and read; a cited answer over the same files gives you the claim and the evidence in one step - which is exactly what you want when the stack is contracts or invoices someone will act on. The workflow is covered end to end in how to chat with multiple PDFs.

Which route for which job