PDF Extractor
PDF extractor pulls structured fields out of invoices, receipts, contracts, resumes, and similar documents using a schema you choose or describe. It is for documents you are allowed to process. Output still needs a human check: scanned tables and odd layouts get misread. The tool uses 1 credit per page. New accounts include a free trial allowance (3 credits by default).
How to use the PDF Extractor
- Upload a PDF you can share for processing. The file is sent so it can be read. Do not upload documents you are not permitted to process.
- Pick or describe the schema. Start from a template, or say which fields you need. Asking for a few specific fields works better than “extract everything”.
- Review the JSON or table. Compare totals and dates with the PDF. Fix names the model merged or missed before you import the data.
When it helps
- Invoice capture. Get supplier, date, and total into JSON so a spreadsheet or script does not start from a blank page.
- A pile of similar forms. Use one schema across a batch, then spot-check the rows that look incomplete.
Common questions
How are credits charged?
1 credit per page. A 4-page invoice costs 4 credits. New accounts include a free trial allowance of 3 credits by default.
Is the extraction guaranteed accurate?
No. Treat it as a draft. Scans, handwriting, and multi-column layouts are the usual failure cases.
Can it crack a locked PDF?
No. Open the file with the password remover only if you already know the password and you are allowed to unlock it.
What do I get back?
Structured data for the schema you asked for, which you can copy as JSON or move into a sheet.
How is this different from the bank statement converter?
The bank statement converter is aimed at statement tables and CSV or Excel output. This extractor is for arbitrary schemas such as invoices and resumes.