Does Adobe Acrobat Do OCR? What It Actually Returns

Last updated July 2026.

Yes. Adobe Acrobat does OCR, through a feature called Scan & OCR. Acrobat Pro supports every recognition mode including fully editable text, Acrobat Standard supports searchable-image output but cannot convert a scan into editable text, and the free Acrobat Reader has no OCR at all. What Acrobat returns is readable text, not accounting fields.

That last sentence is the one that costs people a week. Acrobat will happily turn a scanned invoice into a searchable PDF or an Excel workbook. It will not tell you which number on the page is the sales tax. If you came here because you have a folder of scanned bills and a spreadsheet to fill, read to the end, because the tool and the job are not the same shape.

Which Adobe Acrobat versions include OCR?

Adobe documents three OCR output modes: Searchable Image, Searchable Image and Text, and Editable Text and Images. The tier you own decides which of them you can run.

The Standard limitation catches a lot of buyers. Standard will make your scan searchable, so you can find a word inside it. It will not hand you a document whose text you can rewrite. If editing scanned text is the reason you are buying Acrobat, Standard is the wrong tier, and Adobe says so in its own support documentation.

How do I OCR a PDF in Adobe Acrobat?

Open the file, then go to All tools, choose Scan & OCR, and select In this file. Pick the page range and the document language, then choose Recognize Text. Acrobat builds a text layer behind the image of the page so the words become searchable and selectable. The Settings button in that dialog exposes the output style and the downsampling options, which is where you choose between searchable image and editable text.

One habit worth forming: set the language before you run recognition, not after. Acrobat cannot re-derive characters it never recognized, so a French or Japanese document processed as English gives you a page of confident nonsense that looks fine until you search it.

Does the free version of Adobe do OCR?

Acrobat Reader, the free desktop application most people already have, does not include Scan & OCR. Adobe does publish a free browser tool that recognizes text in a PDF, and it works for the occasional one-page scan. What it does not do is batch, and it does not give you the recognition settings, the language selection, or the suspect-word review that the desktop application exposes. For a single document it is genuinely fine. For a filing cabinet it is a way to spend an afternoon.

Why is OCR greyed out in Adobe Acrobat?

Usually because the page does not need it, or the file will not allow it. If the PDF was generated digitally it already contains a text layer, so there is nothing for recognition to add. If the document is password protected or has editing restrictions applied, Acrobat will not modify it. And Adobe documents a resolution floor: OCR will not be applied to source images below 72 dpi. A photograph downscaled by an email client can drop under that line without looking any different on screen.

How accurate is Adobe Acrobat OCR?

Adobe does not publish an accuracy percentage, which is more honest than most of the industry. What it publishes instead is guidance that tells you where accuracy comes from. The source scan should be 72 dpi or higher for OCR to run at all, 300 dpi produces the best text for conversion, and at 150 dpi accuracy is slightly lower. When Acrobat is unsure about a word it marks it as a suspect, keeping the original bitmap visible with the searchable text behind it so you can review and correct it.

Read that as the real rule of OCR generally: the scanner matters more than the software. A flatbed at 300 dpi and a phone photograph taken at an angle in a parking lot are not the same input, and no recognition engine closes that gap. Handwriting is the outer edge. Adobe markets OCR support for handwritten notes while also warning that handwriting reduces accuracy, and both of those things are true at once.

Can Adobe Acrobat export a scanned PDF to Excel?

Yes, and it runs OCR automatically to do it. Adobe states that Acrobat performs OCR on PDF files containing images, vector art, or hidden text, which includes PDFs created from scanned documents, and you can set the recognition language in the export settings. What arrives in Excel is a workbook that preserves the table layout of the page. The Excel Workbook Settings let you choose one worksheet per table, one per page, or a single sheet for the whole document, and the Numeric Settings control which characters Acrobat treats as decimal and thousands separators.

Set those numeric separators correctly on the way in. It is the single most common reason a freshly exported table shows a total of zero: every value came across as text rather than as a number, and SUM has nothing to add.

Does Adobe Acrobat extract invoice data?

No, and this is the distinction that decides whether Acrobat is the right purchase. Adobe defines OCR as the conversion of images of text into editable characters so that you can search, correct, and copy the text. Characters, not fields. Nothing in Acrobat's Scan & OCR or Export PDF documentation claims to identify a vendor name, an invoice number, a due date, or a total. It gives you the page, faithfully, in a format you can edit.

The gap shows up the moment you have more than one supplier. Export twenty invoices from twenty vendors and you get twenty differently shaped worksheets, because twenty designers put the total in twenty places. A human then reads each one and copies four values into a ledger. That is data entry with extra steps, and it is exactly the work most people were trying to automate when they searched for OCR in the first place. We covered the underlying distinction in OCR vs IDP: OCR reads, extraction understands.

Adobe Acrobat OCR compared to data extraction

So when should you use Acrobat?

Use it when the document is the thing you want. Digitizing a paper archive into searchable PDFs, making a scanned contract editable, redacting, combining, and comparing files: Acrobat Pro is a strong tool and OCR is a real part of it. ABBYY FineReader occupies the same category with a stronger recognition reputation and broader language coverage, and we compared where that ceiling sits on the ABBYY alternative page. Both stop at the same place.

Use something else when the data is the thing you want. Receipts and vendor bills go through receipt OCR software and invoice OCR software that return vendor, date, tax, total, and the line items in consistent columns from any layout, then export a spreadsheet you can actually import. Mixed PDFs of every kind belong in PDF data extraction. If your scans are bank statements rather than invoices, a tool built to turn the statement into a spreadsheet will beat any generic OCR export, because it knows what a transaction row looks like before it starts reading.

And whatever you use, test on your worst paper rather than your cleanest. Published accuracy numbers, ours included, tend to describe conditions your documents will never meet. We wrote about why in how accurate is receipt OCR. Run a hundred real files, count the rows a human had to fix, and buy from that number.

Ready to see the difference? Upload a scanned invoice and compare what comes back against what Acrobat gives you.