We're excited to announce Box Document Split via API, a new capability that is coming soon that enables organizations to split large, multi-document files into individual files, either manually or intelligently using AI, directly within the Box platform.
What Is Box Document Split?
Box Document Split allows customers to take a single PDF (such as a batch-scanned file containing many individual documents) and split it into separate files, each saved to a designated Box folder. This unlocks the ability to classify and extract metadata from individual documents at scale.
There are two modes of splitting:
- Manual Split: Define exact page ranges to split a document into specific files. Powered by Box Doc Gen.
- Smart Split: Use AI to intelligently identify document boundaries within a file and automatically split it into individual documents. Powered by Box Doc Gen and Box AI.
Why This Matters
Organizations across industries receive large volumes of physical mail, contracts, invoices, and forms that are batch-scanned into a single digital file. Before metadata can be extracted from these documents, they must first be split into individual files. Box Document Split solves this by bringing document splitting natively into the Box platform, enabling a seamless end-to-end workflow: Split → Classify → Extract
Top use cases unlocked:
- 📄 Process thousands of batch-scanned mail items by splitting them into individual documents before metadata extraction
- 🏦 Automate intake workflows for financial services firms receiving multi-page loan packages, account applications, or compliance documents
- 🏥 Split multi-record patient files or insurance claim bundles into individual documents for downstream processing
- 📁 Prepare large document archives for AI-powered classification and metadata extraction at scale
Important Notes
- Only PDF files will be supported at GA. Additional formats (PNG, TIFF, JPEG) are planned for future releases.
- Split documents are saved as individual PDF files in the destination folder specified in the API request.
- This feature is only available via API for now.
- Smart Split consumes Box AI Units (~1 AI Unit per 4 pages), but Manual Split does not.
Availability
Box Document Split via API will be available for Enterprise Advanced plans. Both features need Box Doc Gen to be enabled, while Smart Split also needs AI to be enabled.
👉 Join the community to stay informed: AI Content Workflows User Group