Home › PDF utility › Split by Text

Split by Text

Split by Text divides a document whenever a keyword you specify appears on a page, which turns a continuously scanned bundle into separate files per invoice, letter or record. It reads the PDF text layer with PDF.js to find the boundaries.

How to use Split by Text

  1. Upload the multi-document scan or bundle.
  2. Type the keyword that marks the start of a new document, for example 'Invoice' or a reference prefix.
  3. Run the split; a new file is produced for each block between matches.

Where Split by Text is useful

Formats, algorithms and limits

Boundaries are detected by scanning each page's text for your keyword; a page where the keyword appears starts a new output document. Matching is case-insensitive and works on the embedded text layer, so scanned images without text cannot be split this way.

Troubleshooting and honest limits

Frequently asked questions

Is the keyword match case sensitive?

No. Matching ignores case, so 'invoice' and 'Invoice' both trigger a new document boundary.

Can I use a keyword that spans two lines?

No. Only what appears within a single page's text run is matched, so keep the keyword short and on one line in the source document.

Will this work on scanned pages?

Only if those pages have a text layer, meaning they were produced by OCR or exported digitally. Pure image scans have no text to match.

Are the pages before the first keyword kept?

Yes. Anything preceding the first match becomes its own part, so no page is lost.

Related tools

AI Summarizer Compress PDF Edit PDF Image to PDF Converter Merge PDF Translate PDF

Page written and reviewed by the FreeConvertio engineering team on 2026-10-14. Split by Text runs entirely in your browser — no uploads, no account and no daily limits.