How to OCR a PDF on Mac

The problem with scanned PDFs

You scan a contract, a stack of invoices, or a handwritten letter. Your scanner produces a perfectly good-looking PDF. You open it, and everything appears fine - until you try to do something useful with it. You hit Command+F to search for a name or a date. Nothing. You try to select a paragraph to copy it into an email. Nothing. You attempt to highlight a clause. Still nothing.

That's because your scanned PDF is really just a collection of photographs wrapped in a PDF container. Every page is a flat image. Your Mac can display those images perfectly well, but it has no idea what words are written on them. As far as the operating system is concerned, your 47-page contract might as well be 47 photographs of a sunset.

This is where OCR comes in. If you've ever wondered how to OCR a PDF on Mac and actually make those scanned pages useful, the answer is simpler than you might think - and you don't need to upload anything to the internet to do it.

What is OCR and how does it work?

OCR stands for Optical Character Recognition. It's the technology that looks at an image of text and figures out what letters, words, and sentences are present. Think of it as teaching your computer to read.

When you run OCR on a scanned PDF, the software analyses the image pixels on each page, identifies the shapes of individual characters, groups them into words and lines, and then creates a text layer that maps each recognised word to its exact position on the page. This text layer sits invisibly on top of the original image - like a transparent sheet of searchable text perfectly aligned with the visual content beneath it.

The result is a PDF that looks exactly the same as before. The images haven't changed. The layout is identical. But now your Mac understands what's written on every page. You can search the document with Command+F, select text with your cursor, copy paragraphs to the clipboard, and even run Spotlight searches that find content inside the file.

Modern OCR engines are remarkably accurate. Apple's Vision framework, which powers OCR on macOS, uses machine learning models trained on millions of text samples across dozens of languages. It handles printed text with near-perfect accuracy, and it does a respectable job with handwriting too. It recognises text at different sizes, in different fonts, and at slight angles - the kinds of imperfections you get from real-world scanning.

Method 1: Paradise PDF (recommended)

Paradise PDF uses Apple's Vision framework for OCR, which means all text recognition runs entirely on your Mac. Nothing is uploaded anywhere. No internet connection is required. On Apple Silicon Macs, the Neural Engine hardware-accelerates the recognition for significantly faster processing.

Here's how to OCR a PDF on Mac with Paradise PDF:

1. Open Paradise PDF and select the OCR tool from the sidebar.
2. Add your scanned PDF. Drag it from Finder or click to browse. Paradise PDF analyses the document and identifies pages with image-based content.
3. Select your language. Apple Vision supports dozens of languages and can detect multiple languages automatically. Choose the primary language of your document for best results.
4. Click OCR. Each page is processed through Apple's Vision framework. The recognised text is embedded as an invisible layer behind the page images.
5. Save the searchable PDF. Export to your chosen location. The output file looks identical to the original but now supports text selection, search, and copy-paste.

The entire process typically takes a few seconds per page on an Apple Silicon Mac. A 20-page scanned document might take 15-30 seconds depending on complexity.

Once your PDF is searchable, you can do even more with it. Need to edit the content? Convert it to Word for full editing in Microsoft Word or Pages. Want to pull data from scanned tables? The searchable text layer makes extraction far more reliable. You can learn more about the OCR tool on the OCR PDF on Mac tool page.

Method 2: macOS Preview

This is one of the most common misconceptions about macOS: many people assume Preview can OCR a scanned PDF. It cannot.

Preview is an excellent PDF viewer. It can display scanned documents, add annotations on top of them, fill form fields, and even sign documents. But it does not have any OCR capability. If you open a scanned PDF in Preview, you'll see the pages rendered perfectly - but you won't be able to select, search, or copy any text from the scanned images.

This is surprising because macOS does have powerful text recognition built into the operating system. Live Text in Photos and Safari can recognise text in images. But Apple hasn't connected this capability to Preview for PDF files. You can right-click an image in Photos and copy recognised text, but you can't do the same with a scanned page in Preview.

If you only need to read a scanned PDF, Preview works fine. But if you need to search it, extract text from it, or make it accessible to screen readers, you'll need a dedicated OCR tool.

Method 3: Adobe Acrobat

Adobe Acrobat Pro has included OCR for years, and it works reasonably well. The feature is called "Scan & OCR" in modern versions of Acrobat, and it can recognise text in scanned PDFs across many languages.

The main drawbacks are cost and speed. Acrobat Pro requires a subscription starting at around $20 per month (or roughly $240 per year). That's a significant ongoing expense if you only need OCR occasionally. The subscription also bundles a large number of features you may never use - you're paying for a full-featured PDF editor when all you need is text recognition.

Processing speed can also be an issue, particularly with large documents. Acrobat uses its own recognition engine rather than Apple's optimised Vision framework, so it doesn't take advantage of the Neural Engine on Apple Silicon Macs. A 50-page scanned document can take noticeably longer to process compared to tools that leverage the hardware acceleration built into modern Macs.

Acrobat is a solid choice if you already have a Creative Cloud subscription and need OCR as part of a broader PDF editing workflow. But if OCR is your primary need, the subscription model makes it an expensive solution for what is fundamentally a straightforward task.

Method 4: Online OCR tools

A quick search for "OCR PDF" turns up dozens of free online tools: SmallPDF, iLovePDF, OnlineOCR, and many others. They all work the same way: you upload your scanned PDF to a remote server, the server processes it, and you download the result.

For non-sensitive documents, these tools can work in a pinch. But think about what you're actually doing. You're uploading the full contents of your scanned documents to a third-party server operated by a company you probably know nothing about. That server processes your pages, which means your document content exists on their infrastructure, at least temporarily.

Now think about the kinds of documents people typically scan: tax returns with Social Security numbers, medical records with diagnoses and treatment details, signed contracts with confidential terms, identity documents like passports and driving licences, bank statements with account numbers, and legal correspondence covered by attorney-client privilege.

These are exactly the documents where privacy matters most, and they're exactly the documents you'd be uploading to a free online service. Most of these services have vague privacy policies, may retain your files for unspecified periods, and are funded by advertising or upselling premium tiers. Some have been caught retaining uploaded documents far longer than advertised.

Beyond privacy, online tools also have practical limitations. File size limits are common (usually 10-25 MB on free tiers). Processing is slower because of upload and download times. And you need an internet connection, which rules out working on a train, a plane, or anywhere with unreliable connectivity.

Comparison: OCR tools for Mac

Here's how the main options stack up for OCR on Mac:

What to do after OCR

Making a scanned PDF searchable is often just the first step. Once the text layer is in place, a range of workflows open up:

Search and find

The most immediate benefit. Press Command+F in any PDF viewer and search for names, dates, amounts, or any text in the document. Spotlight will also index the content, so you can find the document by searching from your desktop.

Convert to Word or Excel

A searchable PDF converts far more accurately to editable formats than a scanned image does. Use PDF to Word to get an editable document you can revise in Microsoft Word or Pages. If your scanned document contains tables with data, PDF to Excel can extract structured data into a spreadsheet.

Extract and copy text

Select any portion of text and copy it. Quote a clause from a contract, pull an address from a letter, or grab a reference number from an invoice - without retyping anything.

Improve accessibility

Screen readers cannot interpret image-based PDFs. Adding a text layer through OCR makes scanned documents accessible to people using VoiceOver and other assistive technologies. This is also important for compliance with accessibility standards in many workplaces and industries.

Organise and merge

Once your scanned documents are searchable, you can more easily organise them. Merge related PDFs into a single file, split large scans into individual documents, or rearrange pages into the order you need.

Tips for better OCR results

OCR accuracy depends heavily on the quality of the scanned image. The cleaner and sharper the scan, the more accurate the text recognition. Here are some practical tips:

Scan at 300 DPI or higher

The single most important factor for OCR accuracy is resolution. 300 DPI (dots per inch) is the standard recommendation for text documents. At this resolution, individual characters are rendered with enough detail for the OCR engine to distinguish between similar-looking letters. Lower resolutions (150 DPI or below) produce noticeably more recognition errors, especially with smaller font sizes.

Ensure good contrast

OCR works by distinguishing dark text from a light background. Documents with faded text, yellowed paper, or low contrast between ink and page will produce more errors. If your scanner has contrast or brightness adjustments, use them to ensure crisp, dark text on a clean white background.

Straighten pages before scanning

Skewed text - pages that went through the scanner slightly crooked - reduces OCR accuracy. Most modern scanners have auto-deskew features that straighten pages automatically. If yours doesn't, take care to align pages squarely on the scanner glass or in the document feeder.

Clean the scanner glass

Dust, fingerprints, and smudges on the scanner glass appear as marks on every scanned page. These artefacts can confuse OCR engines, particularly when they appear near or over text. A quick wipe with a microfibre cloth before scanning makes a measurable difference.

Use colour or greyscale, not black and white

Scanning in pure black and white (1-bit) mode can clip fine details from characters, especially with serif fonts or small text. Greyscale or colour scanning preserves the subtle gradations that help OCR engines distinguish character shapes. The file will be larger, but the recognition accuracy will be better.

Related reading

• make scanned PDFs searchable
• OCR PDF tool
• combine scanned documents