# Use-Cases - Document AI - Miscellaneous (Splitting & Comparison) - GUC

> How CloudFiles automatically splits merged PDFs into individual, named files and compares document versions to show exactly what changed.

*Document AI for Salesforce · Splitting & Comparison*

## Document splitting and comparison

CloudFiles finds where one document ends and the next begins inside a merged PDF, and separates them into individual, properly named files, then compares any two versions of a document to show exactly what changed between them.

[Book a Demo](https://meet.cloudfiles.io/salesforce?utm_content=sf-document-ai-use-cases-splitting-and-comparison-hero)

*Loved and trusted by leading companies*
Logos: General Mills, Dun & Bradstreet, Wingstop, PenFed, Aviva, Ohio University, CA Conservation Corps, NC State Auditor, Arthrex, Commvault, Computacenter, Aggreko, ITW, Lummus, Northumbrian Water, Hamptons, Anteriad, BrightNight

*What splitting and comparison each do*

## One merged file becomes documents you can actually work with

Splitting takes one large file and separates it into individual documents. Comparison takes two versions of one document and shows what changed.

### Without splitting and comparison

- A batch of invoices or contracts scanned together sits as one long file, so someone opens it and cuts it apart by hand.
- Split files get generic names, so finding the right one means opening several.
- Comparing two versions of a contract means reading both side by side, line by line, by hand.

### With CloudFiles

- CloudFiles finds where one document ends and the next begins, and separates them automatically, each one OCR-enhanced and searchable.
- Each split file is named from what's actually in it, an invoice number, a vendor name, not a generic sequence number.
- Two versions of a document get compared automatically. Added clauses, removed sections, and edited terms surface in seconds.

*Where teams use it*

## Top document processing use cases

How teams already put splitting and comparison to work inside Salesforce.

- **Intelligent Document Splitting** — Automatically find the boundary between documents inside one merged file and separate them, each output file OCR-enhanced and searchable, not a raw image copy.
- **Content-Based Auto-Naming** — Name each split file from what's actually in it, an invoice number, a vendor name, a form type, instead of a generic sequence number.
- **Intelligent Document Comparison** — Compare two versions of a contract, policy, or form and surface what changed: added clauses, removed sections, edited terms, in seconds instead of a manual line-by-line read.
- **Downstream Automation on Each Result** — Splitting a file and finishing a comparison are both events CloudFiles can publish, so a Salesforce flow can act on each new document, or each detected change, without anyone checking back manually.

*How this fits with the rest of Document AI*

## Splitting comes first. Comparison is intelligence at work.

- **[AI-Powered OCR](/salesforce/document-ai/use-cases/ai-powered-ocr)** — Every file that comes out of a split is OCR-enhanced during the process, so what you get is searchable text, not a raw scanned image cut into pieces.
- **[Document Intelligence](/salesforce/document-ai/use-cases/document-intelligence)** — Comparison is a document intelligence capability applied to two versions of one document instead of two different documents.
- **[Custom Document Approval Workflows](/salesforce/document-ai/use-cases/custom-document-approval-workflows)** — Once a bundle is split, each individual file can carry its own checklist and its own approval path, instead of one review covering a mixed, merged file.

*Why teams pick CloudFiles for splitting and comparison*

## Output you can act on immediately

- **Searchable from the start** — Every split file is OCR-enhanced during the process, so what you get is searchable text, not a raw scanned image cut into pieces.
- **Named from real content** — Files are named from what's actually inside them, an invoice number, a vendor name, a form type, instead of a generic sequence number.
- **Triggers downstream automation** — Splitting a file and finishing a comparison are both events CloudFiles publishes, so a Salesforce flow can act on the result without anyone checking back manually.
- **Works across your storage** — It runs on files already in Salesforce and on files in a connected external drive, not only one or the other.

## How intelligent PDF splitting actually works

Splitting runs in three steps: CloudFiles finds the logical boundary between documents inside a merged file, OCR-enhances each page so scanned or image-based content becomes searchable text, then names each output file from what's actually in it, an invoice number or a vendor name. You can also give it instructions up front for documents with an unusual structure, to improve accuracy.

[Read the help doc](https://help.cloudfiles.io/salesforce/intelligent-pdf-splitting)

## Stay compliant & secure.

CloudFiles is independently audited for SOC 2 Type II, certified to ISO 27001, and compliant with HIPAA and GDPR. Data residency is supported in the US, EU, UK, and AU. Your files stay in your own storage, CloudFiles never re-hosts them, and are not used to train our AI models.

Certifications: SOC 2 (Type II), HIPAA (Compliant), ISO 27001 (Certified), GDPR (Compliant)

[Security overview](/security)

## Your customers won't wait.

Book a demo and see a merged file split, or two versions compared, in real time.

[Book a Demo](https://meet.cloudfiles.io/salesforce?utm_content=sf-document-ai-use-cases-splitting-and-comparison-footer)

## Splitting and comparison, answered directly

**Q: What does intelligent document splitting actually do?**
It looks inside one merged file, a batch of scanned invoices or contracts saved as a single PDF, and finds where each individual document starts and ends, then saves each one as its own file.

**Q: How are the split files named?**
From their own content, an invoice number or vendor name pulled out of the page, not a generic "Document_1" label.

**Q: Does splitting work on scanned, image-only PDFs?**
Yes. Each split file is OCR-enhanced during the process, so the output is searchable text, not just a smaller image file.

**Q: What does document comparison do, exactly?**
It takes two versions of the same document, a redlined contract, an updated policy, and identifies what changed between them: additions, removals, and edits.

**Q: Can something in Salesforce react automatically once a document is split or compared?**
Yes. Both splitting and comparison completing are events CloudFiles publishes, so a Salesforce flow can trigger off either one without anyone checking back manually.

**Q: Does this only work on files already in Salesforce?**
No. It works on files in Salesforce and on files in a connected external drive.