AI · August 14, 2026 · 6 min
How to extract fields and values from a PDF form
Turn a filled application or intake packet into structured JSON without retyping every box.
Open Extract Form Fields- 01
Open Extract Form Fields and drop the filled PDF.
- 02
If it may be a scan, turn on AI OCR if a scan. OCR runs only when local text extraction is empty and reads at most the first two pages.
- 03
Review the labels and values the model returns.
- 04
Download JSON if you need to pipe it into another tool.
Re-keying a 12-page intake is how errors get into the CRM. Extraction is faster — if you treat it as a draft, not gospel.
Labels matter more than pixels
A form extractor needs “Date of birth” next to a value. Letterhead art does not help. PDF Boost AI sends extracted text, or — only when you opt into AI OCR and the text layer is empty — downscaled images of the first two pages. OCR and form extraction count as separate AI runs.
Check the empty boxes
Unchecked fields and “see attached” notes are where automation lies. If a value looks too neat, open the page.