Extract Data from PDF into One Report

Extracting data from a PDF by copy and paste is slow, and the document's properties stay hidden inside the file. This tool reads a PDF and writes one report with its metadata, its full text and a list of the dates, email addresses, phone numbers and money amounts it found. It's handy for a quick look at what an invoice, report or form holds before you start working with the data.

What the PDF data extraction report contains

The report is a PDF with three sections. The first lists the document's metadata: page count, title, author, subject, keywords, creator, producer, creation date and modification date. Empty fields show a dash, since the software that made a PDF doesn't always fill them in.

The second section, Detected Fields, lists every date, email address, phone number and currency amount in the text, each one once. The third holds the extracted text itself, line by line.

Which dates, phone numbers and amounts are detected

Dates get picked up when they're written like March 14, 2025 (with the English month name), 2025-03-14 or 14/03/2025. Phone numbers only count in the ten-digit North American style, such as (415) 555-0199, with an optional country code. Amounts need a $, € or £ sign in front or USD, EUR or GBP after them, so a bare number like 1,249.50 won't be listed.

Does it read filled-in form fields and European formats?

Form values only come through when the form shows them as page content, and European dates, amounts and phone numbers aren't detected. We tested both in October 2026.

We filled three form fields with a name, the date 2026-03-14 and $120.00. When the fields carried their own appearance (which is how most PDF editors save them), and when the form was flattened, the report listed all three values and detected the date and amount. With the same values stored only as field data, we got the labels and no values at all.

So if a filled form comes out blank, run it through Flatten PDF, one of our tools to optimize PDFs, then extract.

On a page with ten values in different national styles, Detected Fields caught just 4 of the 10. The extracted text below it still had all ten lines, unchanged.

How to extract data from a PDF

1. Upload one PDF, or drag it into the upload area.
2. Select Convert To PDF to start the extraction.
3. The results screen lists the dates, email addresses, phone numbers and money amounts it found, each with a Copy button. Download saves the full report as a PDF named after your file, ending in -extracted.pdf.

The tool reads one PDF per run, so upload several files one after another. Your original file stays as it is.

How the text is read from the PDF

The text comes from the PDF's text layer, read with Ghostscript's txtwrite device, which outputs the text contained in a document as Unicode. The metadata comes from the pdf-lib library. You get the text line by line, without the original fonts, colors or images.

Text PDFs and scanned PDFs

A PDF made from a word processor or a web page has a text layer, so its words come through. A scanned page is usually just a picture, and then the report shows little or no text. Run a file like that through our OCR PDF tool first, since optical character recognition converts images of printed text into machine-encoded text.

When a different PDF tool fits better

The report is meant for reading and checking. It won't load into a spreadsheet, so for tables use our PDF table extractor, which writes CSV, or convert the whole file with PDF to Excel. For plain text you can edit, use PDF to TXT, and to change the title, author or other properties, use Edit Metadata.

How to check the extracted data

Check important numbers, names and dates in the report against the source before you rely on them. A pattern can also catch a number that only looks like a date or a phone number, so treat Detected Fields as a starting point. Metadata is whatever the creating software wrote, which means it can be missing or out of date.

The tool takes PDFs up to 25 MB and you don't need an account. We delete your upload and the server's working files as soon as the report has gone out.

What happened when we tested PDF Data Extraction

We ran pdf-data-fields-fixture.pdf, our own one-page test PDF of 975 B with an invoice date, an email address, a phone number and a total due. There was nothing to set, and the report reached the browser 0.2 seconds after we pressed the button.

The report, pdf-data-fields-fixture-extracted.pdf, is a one-page PDF of 12.7 KB. Tested in October 2026 on a local copy of our server.

Frequently Asked Questions about Extracting Data from PDF

What does the PDF data extraction report contain?

The document's metadata, a list of detected dates, email addresses, phone numbers and currency amounts, and the full extracted text.

Can the tool extract text from a scanned PDF?

Only if the PDF has a text layer. A scan made of page images needs OCR first, for example with our OCR PDF tool.

Will the original PDF be changed?

No. The report is a new PDF, and your original stays untouched.

Can it extract tables into a spreadsheet?

No. Table text appears line by line in the report. For CSV output, use PDF Table Extract.

Can I process several PDFs at once?

No. The tool reads one PDF per run, so upload the files one after another.

Is my PDF stored on your server?

No. We delete the upload and the working files as soon as the report has gone out.

Do I need to install software or create an account?

No. It works in any modern browser, with nothing to install and no sign-up.