How to Rename PDF Files: 5 Methods Compared (2026)

Every scanner creates the same problem: files named scan_001.pdf, Document.pdf or IMG_20260109.pdf. There are five ways to fix that, from a free keystroke to software that reads the page and writes INV-2026-0115_Acme-Corp.pdf on its own. Which one fits depends on two questions: do your files share one layout, and are several documents stuck together in one PDF? Uniform files from one source call for a batch utility; mixed scans with several documents per file call for content-based renaming; everything in between is settled by the comparison table below.

Why file names matter

You're looking for the Acme Corp invoice from January 2026.

Without naming:                      With naming:
scan_001.pdf                         INV-2026-0115_Acme-Corp.pdf   ← found
scan_002.pdf
...
scan_047.pdf  ← this one? open it...

Worked example, assumptions only: if a search through unnamed files costs about five minutes and you run ten of them a month, that's nearly an hour a month per person. With named files the same search is a glance at the folder. Your numbers will differ; the direction won't.

Method 1: By hand

Windows: select the file, press F2, type the new name, Enter, repeat. Mac: select, press Enter, type, repeat.

Free, full control, no software. The downsides are the ones you'd expect: it's slow, every colleague invents their own scheme, typos creep in, and because it's dull it gets skipped, which is how the backlog folder is born. Reasonable below about ten files a month.

Method 2: Batch utilities (PowerRename, Bulk Rename Utility, Finder)

Windows: PowerToys PowerRename (free) adds search-and-replace with regex to the right-click menu; Bulk Rename Utility goes further. Mac: select several files, right-click, Rename, and Finder offers replace-text, add-text and counter formats.

These tools shine when the names already follow a pattern: IMG_2026XXXX.pdf becomes Receipt_2026XXXX.pdf in one preview and one click, hundreds of files at a time. Their limit is just as clear: they transform the old name, they never look inside the file. A folder of scan_001.pdf to scan_047.pdf contains no information to transform, so batch utilities can't help there.

Method 3: Adobe Acrobat Action Wizard

Acrobat Pro (paid subscription, see Adobe's pricing page) includes an Action Wizard for JavaScript-based batch actions. You define text-extraction regions or regex patterns per document type, and the script renames files from the extracted fields.

It works offline and costs nothing per file once set up, but the setup is real: scripting knowledge, 2 to 4 hours per document type, and a separate action for invoices, contracts and every other layout. Scripts break when a vendor redesigns their invoice, and scanned PDFs need an OCR pass first because the wizard works on text coordinates, not pixels.

Method 4: Template-based OCR renamers

Tools such as Win2PDF's Auto-name feature, or DMS workflows in paperless-ngx, run OCR and build the filename from zones or matched fields that you define per document type. Set up once per layout, they rename reliably for as long as the layout holds.

The economics work when the same sender mails the same format every month. A mixed stack of 20 senders means 20 templates, and a batch scan with several documents inside one PDF has to be split before these tools can touch it.

Method 5: AI that reads the content

Docusplit sends each page as an image to a vision model (GPT-4.1 through the OpenAI API). There's no OCR step and no template: the model reads the page the way you would.

What it extracts: in invoice mode the invoice number, company and invoice date; in document mode first whether the page is an invoice, then either those invoice fields or a document type, sender and date. Amounts, tax IDs and bank details are not read.

Input:  scan_047.pdf

AI reads:  invoice? yes · company: Acme Corp ·
           number: INV-2026-0115 · date: 2026-01-15

Output: INV-2026-0115_Acme-Corp.pdf

A letter in the same stack becomes Letter_Northwind-Ltd_2026-01-20.pdf. Because the model reads content rather than coordinates, layout changes don't break anything, and a 50-page PDF containing 25 different documents is split and named in the same run.

The trade-offs: it's a paid cloud service beyond the free 30 pages, processing takes about 6 to 10 seconds per page rather than being instantaneous, the naming scheme is fixed rather than customizable, and unreadable pages fall back to Document_Page_N.pdf for you to fix by hand.

The five methods side by side

Method 1 is you. Methods 2 to 4 rename by rules you write (patterns, text positions, templates) and stop working when the layout changes. Method 5 reads the page and needs no rule, which is also why it's the one that can take a mixed 50-page scan and return individually named files.

Which method fits your situation

• Names already follow a pattern: method 2.
• You have Acrobat Pro, JavaScript skills and one stable layout: method 3.
• Same vendor, same layout, every month, high volume: method 4.
• Mixed senders, scans, several documents per PDF: method 5.
• Fewer than ten files a month: method 1 does the job.

What it costs, worked example

Assumptions, not measurements: 50 mixed files a month, about 30 seconds each to open, read and rename by hand, makes 25 minutes. At $50 an hour that's roughly $21 of your time each month. The Starter plan costs €9.99 for 100 pages, so on these assumptions the break-even sits near 25 mixed files a month; below that, F2 wins. Run your own figures in the ROI calculator.

The naming convention worth adopting

[DocumentType]_[Sender]_[Date].pdf

Contract_Northwind-Ltd_2026-01-15.pdf
Reminder_City-Utilities_2026-02-20.pdf

Type first for quick visual filtering, sender second because it's what you search for, date last in ISO form (YYYY-MM-DD) so files sort chronologically. Two variants serve special cases: Date_Type_Sender when chronology matters most, Sender_Type_Date inside per-vendor folders.

Docusplit applies this scheme automatically, with one deliberate exception: invoices are named InvoiceNumber_Company.pdf (for example INV-2026-0115_Acme-Corp.pdf), because the number is what accounting searches for. Names are capped at 100 characters, spaces become underscores, and special characters are stripped. If you name files by hand, the same rules serve you well: no spaces, no version suffixes like final_v2, one consistent scheme.

Renaming and splitting in one pass

Most tools do one of the two jobs. iLovePDF and Smallpdf split PDFs, and iLovePDF's Smart Range mode (since 2026) even finds the document boundaries with AI, but the parts come out without content-based names. Single-file AI renamers name but don't split. Acrobat can script both with real effort. Few tools do both in one pass without templates; that's the gap Docusplit targets: upload a 50-page scan holding 25 documents, get 25 named PDFs back in a ZIP after roughly five to eight minutes, plus a CSV listing what was read from each one.

What typically goes wrong

Handwriting, faint thermal receipts and scans below roughly 200 dpi produce Document_Page_N.pdf fallbacks with "Not detected" in the CSV rather than guessed names. Company names can come out in a variant spelling, and an attachment without a letterhead can end up under the wrong document. There is no correction step inside the tool; the CSV shows you which files to fix, and for a typical batch that's a minute of hand-renaming. The full failure list lives in the document separation guide.

FAQ: renaming PDFs by content

Can I rename PDFs based on content for free?

Up to a point, yes: Docusplit renames 30 pages for free after you create an account, no credit card needed. It reads each page and names the file from its content. Beyond that, plans start at €9.99 a month for 100 pages.

Does AI PDF renaming work with scanned documents?

Yes. The pages are analyzed as images by a vision model, so a scan and a born-digital PDF are treated the same and no separate OCR step is needed. 200 dpi scans work, 300 dpi is the safe choice; handwriting and very faint receipts are the known exceptions.

How accurate is AI-powered PDF renaming?

There is no measured accuracy figure to quote. What you get instead is a predictable fallback: a page the AI can't read is named Document_Page_7.pdf and listed as "Not detected" in the CSV, never silently mislabeled. Expect to hand-fix the odd file, typically the handwritten or faded ones.

Can Docusplit rename AND split a multi-page PDF at the same time?

Yes, that's the core feature. Upload one large PDF; the AI detects where each document starts, splits the file, and names every part from its content in the same run.

What happens with multi-page documents?

Pages that continue a document, recognizable by cues like "Page 2 of 3", a missing letterhead or a table that carries on, stay with their first page and share one filename.

Which languages are supported?

The AI reads documents regardless of language; German, English, French, Spanish, Italian and Dutch are the tested ones. The interface and the CSV come in English and German, and document-type labels follow the interface language.

Pick the method that matches your files

Uniform names, use a batch utility today for free. One stable layout in volume, script it once. Mixed scans with several documents per PDF, let the AI read them: the first 30 pages are free after creating an account, no credit card needed. Start for free

Related articles:

• Split scan PDFs into individual documents
• Detect and split invoices
• Auto-organize documents into folders