DocForensic

Split PDF by Text

Split a batch of invoices, statements or payslips wherever a text pattern appears, and name each file from the text next to it.

Tell the tool what marks the start of each document - "Invoice No:", "Statement Period", "Page 1 of" - and it finds every page where that text appears and splits the batch there. Patterns can be plain text, a whole word, a full line or a regular expression, and several can be combined. Each document is named from the text found next to the match, such as the invoice number or account number, with dates tidied into one format.

The text is read once, so the live pattern tester, the page view and the results table update instantly as you change the pattern. Scanned batches without a text layer can be read with OCR. Before splitting you can see every match, where each document starts, and which pages need a second look, and you can insert or remove splits and edit filenames.

Simple workflow

How it works

  1. 1

    Add a batch

    Add one or more PDFs and enter the text that starts each document.

  2. 2

    Find matches & review

    Check every match and document boundary as you refine the pattern, and fix any split by hand.

  3. 3

    Split

    Download each document named from its matched text, or a ZIP with a split log.

Frequently asked questions

Which split rule should I use?

Use "Split before every match" when each document starts with the marker text, such as an invoice number on its first page. Use "Split after every match" when the marker is on the last page, such as "Total due". Use "Split when the captured value changes" when every page carries the same ID, such as an account number on each page of a statement.

My pattern finds nothing - what should I check?

Try turning case sensitivity off, use "Contains" instead of an exact line, and make sure "Ignore extra whitespace" is on. If the pages are scans, they have no text layer yet: turn on OCR in the settings.

How do I name files from the invoice number?

Pick the "Invoice number" preset, or write a regular expression with a capture group around the number and choose "Regex capture group" under Value Capture. The captured text fills the {captured} token in the filename pattern.

Stay ahead of your document work

New tools and thoughtful product updates, delivered occasionally. No noise, no spam.