How to Batch Rename Scanned Documents with OCR on Mac

A scan backlog looks the same on every Mac. A flatbed scanner or multifunction printer hands you scan.jpg, scan 1.jpg, scan 2.jpg, incrementing forever. Scan a document from your iPhone straight into a Mac folder and it lands as Scanned Document.pdf, then Scanned Document 2.pdf. A photographed page from a scanning app saves under whatever generic name the app defaults to. None of it says whether the file is a lease, a lab result, or last year's W-2.

This guide covers what OCR (optical character recognition) is actually doing when you scan a document, what macOS can already do with that text natively, where those native tools stop being enough once you have more than a handful of scans, and the batch workflow in Magic Rename for clearing out a folder of them at once.

What OCR Actually Does (and Doesn't Do)

A scanned page, whatever produced it, a flatbed scanner, a phone camera, a multifunction printer, is just a picture. To a Mac, a scan of a lease and a scan of a grocery list look identical: a grid of pixels with no idea what any of them mean. OCR is the step that looks at those pixels and works out which shapes are letters, turning a photograph of text into text a computer can actually read, select, and search.

That's the whole job, though. OCR on its own hands back a block of extracted text; it doesn't know that the third line is a vendor name and the date near the top is the one that matters for a filename. Something else has to read that extracted text the way a person would, pick out the useful fields, and turn them into a name. That's the gap between "my scan is now searchable" and "my scan is now named something useful," and it's where the native tools below stop short.

What macOS Does Natively: Live Text, Embed Text, and Shortcuts

macOS has real OCR built in, it's just spread across a few different features, each solving a narrower problem than "rename my scans."

Live Text

Live Text lets you hold the pointer over text inside a photo or a scanned page in Preview, drag to select it, and copy, look up, translate, or search it, the same way you'd select text on a webpage. It's genuinely useful for pulling one phone number or one line out of a scan you already have open. It has no batch mode, no folder-wide action, and no rename step; it operates on whatever single image is in front of you.

Preview's Embed Text export

Since macOS Sonoma, Preview can run OCR on a scanned PDF through File > Export, choosing Embed Text. It recognizes the text on the page and embeds it in the PDF so the document becomes selectable and shows up in Spotlight searches, the option only appears if the PDF doesn't already have a text layer. It's a real OCR feature, but it works on one file at a time through a manual export dialog, and it stops at making the document searchable. The filename stays exactly what it was.

Shortcuts: Extract Text from Image + Rename File

For an actual rename, Shortcuts has an Extract Text from Image action built on Apple's OCR engine, which you can chain into a separate Rename File action to build a basic content-based renamer entirely from stock macOS. This is the closest native option to what this guide is actually about, and it's covered in full, with the exact action chain, in the PDF content-renaming guide.

Where Native OCR Renaming Breaks Down at Scale

A Shortcuts OCR-to-rename pipeline has no concept of "this text is the date" versus "this text is the sender," it just extracts raw text in whatever order it sits on the page. That's fine for a stack of identical monthly statements from one bank. It falls apart the moment a backlog mixes a flatbed-scanned lease, a phone-photographed receipt, and a Continuity Camera scan of a form, three different layouts, three different places the useful information happens to sit.

Format is a second wall. Live Text and Embed Text both work on images and PDFs individually, but nothing in Preview treats a folder of mixed .jpg, .tiff, and .pdf scans as one job. Clearing a real backlog means opening each file, running the export, and repeating it by hand, which is exactly the kind of task that doesn't scale past a handful of documents.

An AI-based reader closes both gaps at once: it understands the extracted text semantically rather than positionally, so it can tell a signing date from a document title even when they land in different spots on different scans, and it treats a mixed folder of formats as a single batch. That's the problem Magic Rename is built to cover.

Batch Renaming Scanned Documents, Step by Step

Here's the workflow in Files Magic AI, from a folder of generic scan names to one that's searchable at a glance.

Good vs. Bad Filenames for Scanned Documents

Where Scans Actually Come From

"Scanned document" covers more sources than people usually count, and each one has its own default naming habit that turns into the same problem later.

• Flatbed scanners and multifunction printers, driven through macOS's Image Capture app, default to naming scans scan.jpg, then scan 1.jpg, scan 2.jpg, incrementing with every save unless combined into one document.
• Continuity Camera lets you scan a document from a nearby iPhone or iPad directly into a Mac folder from Finder's contextual menu; the resulting PDF is named Scanned Document.pdf, then Scanned Document 2.pdf, and so on.
• Third-party scanning apps on iPhone save their own generic defaults, and often export as PDF with no field extracted from the page itself.
• A photo of a page, taken with an ordinary camera app rather than a scanning app, lands in the camera roll under a standard IMG_XXXX name and is functionally a scan for renaming purposes, even though nothing ever called it one.

Magic Rename treats all four the same way: read what's actually on the page, propose a name from it, and let the source app's naming habit stop mattering.

Native Tools vs. Magic Rename, Side by Side

None of the native tools are bad at what they specifically do. Live Text is the fastest way to copy one phone number out of a scan you have open. Embed Text is the right call for making a single important PDF searchable. Shortcuts is genuinely enough for a folder that only ever holds one recurring form. What none of them do is treat a mixed backlog of scans as one job with a filename at the end of it, which is the specific gap this guide is about.

Sources & Further Reading

Frequently Asked Questions

A scanner has no idea what it just scanned, so it names the file after itself instead: the app, a counter, a timestamp. OCR is what makes the actual content readable again, and reading it well enough to name the file is the part native macOS tools split across several single-file features rather than offering as one batch job. For documents that arrived as PDFs specifically, with a text layer already or without one, the deeper walkthrough is in renaming PDFs by content. For the bookkeeping-specific case of scanned receipts, see renaming scanned receipts on Mac, and for scanned invoices, how to rename invoices on Mac. For the full picture of how AI renaming engines work, including on-device versus cloud processing, read How AI File Renaming Works on Mac.

Published August 22, 2026 · More guides · Files Magic AI