Video Tutorials

What is Fact Extract?

Get a one-minute overview of Fact Extract, a system that reads collections of documents and organizes the facts they contain into spreadsheets and searchable databases. You can upload PDFs to FactExtract.net and receive the high quality AI OCR, identification of individual documents within large PDFs, and tailored summaries. View the results in Excel or our free Fact Extract Desktop app. Search, extract files, organize notes, and generate timelines. Optionally connect to AI with Claude, Goose, or AnythingLLM to ask questions in plain English and receive answers with clickable citations to the supporting PDF pages.

Meet Fact Extract

Get a three-minute introduction to the complete Fact Extract system. There are two ways to begin: build a free local database from the files on your PC, or upload PDFs to FactExtract.net for the full treatment, with simple pay-by-the-page pricing. AI vision reads difficult pages such as faded carbons and photostats in multiple passes, finds where each document begins inside a large scanned PDF, and returns an Excel workbook in which every value cites its source document and page. The same database then loads into the free Fact Extract Desktop app for timeline analysis, and connects to Claude for plain-English questions, cited answers, and generated documents such as timeline graphics.

Drag. Drop. Done.

Take a one-minute to see the Fact Extract workflow. Upload a stack of PDFs, pick your questions to ask, and get a spreadsheet and database with superior OCR of your documents, plus a summary. Use our free Fact Extract Desktop app to build timelines and search your documents by keyword or by idea. Connect your AI. Get get answers with links to every cited page, so you can verify the AI got it right.

Private by Design

Get a 45-second summary of how FactExtract.net protects your documents. Uploaded files are deleted the moment processing ends, and no human being ever sees them. The AI processing step keeps nothing: no prompts, no responses, no training on your data. Results are deleted automatically after fourteen days, or immediately with the Delete My Data button. The video shows each protection in the published privacy policy and in the product itself.

The Complete Tour

Ready to learn more? Take a more in depth eight-minute tour of Fact Extract. Follow one set of real documents, Department of Justice FOIA files on Al Capone, from upload at FactExtract.net to finished research. The tour covers choosing a structure, the quote shown before any charge, the automated OCR and extraction pipeline, and the two downloads: an Excel workbook with clickable citations and a database of content and summaries. Learn how the free Fact Extract Desktop app allows powerful searching, note taking, grouping pages into collections, exporting page and documents, and generating timelines - all private and local to your machine. See how you can optionally connect the database to Claude (or other AI models through Goose or AnythingLLM) for plain-English questions with cited answers. Also learn about the robust built-in help system, using AI to save findings for future research, and whole-database exports for tools like NotebookLM or ChatGPT.

What Is a Structure?

Get a four-minute explanation of structures in Fact Extract. A structure is the list of questions you want answered about every document: each question is a field, and each field becomes a column in your finished spreadsheet. Every upload to FactExtract.net starts by picking one. The video shows how documents are analyzed in chunks, how the overview field gives the AI context that improves every answer, and the three ways to create a structure: start from a ready-made template, fill in the form using plain-English questions, or let Claude or another AI write the whole structure from a briefing prompt the app provides. It closes with Customize OCR, which tailors how pages and pictures are read.

Meet Fact Extract Prep

Get an essential overview of Fact Extract Prep in 3 minutes. Convert all the files in a folder (and its subfolders) to PDF. Emails are opened, printed to PDF, and so are their attachments. Videos are converted to a PDF with a page of metadata and then snapshots of the video so you can quickly see its contents. Files that cannot be converted (plus any original video files) are saved in native form. Includes an index in clickable HTML. Use to prepare a large body files for use in Fact Extract Desktop of FactExtract.Net. Also helpful for preparing files for civil discovery, converting a set of folders with files in into PDFs ready to Bates Stamp.

Meet Fact Extract Bookmarker

In 4 minutes, learn how to use Fact Extract Bookmarker. Easily subdivide large PDFs into smaller files at bookmark locations. Add or edit bookmarks by hand using easy keyboard shortcuts, split at existing bookmarks, or use your own AI to find each internal document and bookmark it. Helpful for accurately splitting up Electronic Medical Records (EMR) and civil discovery files. Free and private.