Documentize — The Complete Guide
It is 5 p.m. A 90-page report has to reach a client tonight, and your mail system just bounced it for being too big. This book is how you fix that in a few clicks, and how you stop dreading the next one.
You may know the website: products.documentize.app and products.documentize.ai. Each page there does one small job well. This book shows what each page does, what it will not do, and how to chain pages so that real work gets finished: a big report sent by email, phone photos turned into a searchable PDF, a 90-page study understood before tomorrow's meeting. It is honest about the gaps, and it tries to be fun to read.
Start here, whatever you need
- I have a task right now → open the goal finder, or go straight to a recipe such as send a big report by email.
- I want to know every tool → read Part I, then the concepts, then your family of apps.
- I want to automate → jump to Beyond the browser, after doing one recipe by hand.
How this book is built
- Vision → why small focused apps beat one giant editor (Part I)
- Concepts → the seven ideas every app shares (Part II)
- Map → how all apps relate to each other (Part III)
- Apps → one page per app, same template, grouped by what you want to do (Parts IV–XI)
- Recipes → chains of apps that reach a goal (Part XII)
- Beyond the browser → CLI, MCP, API, self-host (Part XIII)
- Rescue & reference (Parts XIV–XV)
Every path ends back at Part III's map, so you can always zoom out again.
File layout: one file per entry; folder = chapter; index.md = chapter landing page.
Part I — Vision: Why Documentize
01-vision/
- I.1 The problem: documents are a pile of small chores
- Typical day: merge, shrink, unlock, OCR, sign, convert, explain
- Why one big editor is the wrong answer; why many small focused apps are the right one
- I.2 The Documentize idea: one app = one job
- Open, drop file, get result — no install, no required account
- Every app does one thing well, results feed the next app
- I.3 Two sites, one product family
- products.documentize.app — classic PDF tools
- products.documentize.ai — AI tools (understand, generate, enrich)
- Which site to open for which need (decision table)
- I.4 Who this book is for and how to read it
- Path A "I have a task" → Part VIII recipes
- Path B "I want to know every tool" → Parts IV–XI
- Path C "I want to automate" → Part IX
- I.5 Product portfolio at a glance
- All ~50 apps on one page, grouped (generated from Part III map)
Part II — Core Concepts Shared by Every App
02-concepts/
- II.1 The universal flow: Upload → Options → Run → Result
- Anatomy of an app page (upload zone, options panel, run button, result panel)
- Upload methods: drag & drop, file picker, multiple files, URL/cloud where offered
- Result page: download, preview, share link, "open in another app"
- II.2 What a "document" is: formats and file types
- PDF (text-based vs scanned vs image-only), DOCX, images, HTML, SVG, archives, EPUB/MOBI/TXT
- Which formats each app accepts and produces (matrix → Appendix A)
- II.3 Tasks, options and results
- Required vs optional options, sensible defaults, "what changes if I change this"
- Output naming and what happens to your original (never modified)
- II.4 Limits, plans and the daily allowance
- File size, page count, time limit per task, daily operations
- Continuation: finishing a partly-processed batch on another day, free of charge
- Anonymous vs signed-in use
- II.5 Privacy and data lifecycle
- Where files are processed, how long they live, automatic deletion
- What the AI apps send to models and what they do not
- Password-protected and confidential documents: safe-handling rules
- II.6 Language and localisation
- Interface languages vs document languages (OCR, translation, AI output)
- II.7 Quality model: "lossless", "lossy", "re-generated"
- Which apps leave content untouched, which rewrite it, which invent new content (AI)
- Why this matters for order in a chain
Part III — The Map: How Apps Relate
03-map/
- III.1 The app family tree
- Eight families: Organize · Add & Structure · Protect & Prove · Shrink & Repair · Extract & Search · Convert & Create · Understand with AI · AI for Developers
- III.2 The pipeline model: input → state → output
- Every document has a state (scanned? locked? large? has outline? has text layer?)
- Every app needs a state and changes a state
- III.3 App dependency matrix: "needs", "improves", "breaks"
- Example rows: OCR → Searchable → Search; Deskew → OCR; Unlock → almost everything; Compress last, Sign last
- III.4 Golden rules of ordering
- Repair before analyse · Fix orientation before OCR · Edit before sign · Compress after edit · Protect last
- Anti-patterns (what silently destroys signatures, text layers, bookmarks)
- III.5 Similar-looking apps: which one do I pick?
- Compressor vs Optimize · Searchable vs OCR Image vs Extract Text · Add TOC vs TOC AI
- Flatten vs Protect · Remover/Delete Page vs Splitter · Resize vs Compress · Parser vs Extract Form Data
- Sign vs Verify Signature
- III.6 Goal-to-app finder
- "I want to… → use…" lookup (search-friendly, one line per goal)
Part IV — Organize: Pages, Files, Layout
04-organize/ · family overview
Pages in, pages out. These are lossless and the usual first or last step.
- IV.1 Merger — combine many files into one PDF
- IV.2 Splitter — cut one PDF into many (by pages, ranges, size, bookmarks)
- IV.3 Delete Page — remove unwanted pages
- IV.4 Rotation — fix page orientation (all, odd/even, selected)
- IV.5 Resize — change page size (A4/Letter/custom), scale vs fit
- IV.6 Booklet — reorder pages for printing folded booklets
- IV.7 Chapter lab: tidy a messy scan pack
- IV.8 How the family works together
Part V — Add & Structure: Make a Document Navigable and Richer
05-add-structure/ · family overview
- V.1 Add TOC — build a clickable table of contents from rules you set
- V.2 Table of Contents AI — detect headings and hierarchy automatically
- Add TOC vs TOC AI: control vs automation
- V.3 Add Table — insert tables with your data into pages
- V.4 Flatten — bake forms, annotations and layers into the page
- V.5 Collage — compose images/pages into one sheet with AI layout
- V.6 AI Illustrator — generate illustrations for scenes in your text
- V.7 Chapter lab: from raw manuscript to navigable illustrated PDF
- V.8 How the family works together
Part VI — Protect & Prove
06-protect/ · family overview
- VI.1 Protector — lock a PDF with a password (one password to open; what it does and does not restrict)
- VI.2 Unlock — remove a password you know (web: PDF; Office files via the CLI plugin)
- VI.3 Sign — add a digital/e-signature
- VI.4 Verify Signature — check that signatures are intact (a mode of the Sign page)
- VI.5 PDF/A Validator — archive-grade compliance check, what failures mean
- VI.6 PDF Redactor — not available yet: what to do instead
- VI.7 Compare PDFs — not available yet: how to compare two versions manually
- VI.8 Chapter lab: prepare a contract for external sharing
- VI.9 How the family works together
- Order that keeps signatures valid: edit → flatten → compress → protect → sign
Part VII — Shrink, Clean & Repair
07-shrink-repair/ · family overview
- VII.1 Compressor — reduce file size for email/upload (quality levels)
- VII.2 Optimize — structure-level clean-up (web-fast view, duplicates, fonts)
- VII.3 Deskew — straighten crooked scans
- VII.4 Recovery — rescue damaged or half-readable PDFs
- VII.5 Unwatermark — remove watermarks (and when it will not work)
- VII.6 Chapter lab: take a 200 MB phone-scan to a 5 MB, straight, readable PDF
- VII.7 How the family works together
Part VIII — Extract, Read & Search
08-extract/ · family overview
- VIII.1 OCR Image — text from photos and scans
- VIII.2 Searchable — add an invisible text layer to a scanned PDF
- VIII.3 Extract Text — get plain text out (with or without layout)
- VIII.4 Extract Image — pull embedded pictures out
- VIII.5 Extract Metadata — author, dates, producer, custom properties
- VIII.6 Extract Form Data — field names and values from filled forms
- VIII.7 Parser — take a PDF apart into text, images, attachments, bookmarks, annotations and form data
- VIII.8 Search — find text inside a PDF · Global Search — site-wide search box (not a document search)
- VIII.9 Resume Analyzer — extract data and score a CV against a job
- VIII.10 Chapter lab: turn a stack of scans into a searchable, queryable archive
- VIII.11 How the family works together
Part IX — Convert & Create
09-convert-create/ · family overview
- IX.1 Converter — PDF ↔ Word, HTML, images, etc.; what each target preserves
- IX.2 SVG Maker — text-to-layered-vector art with AI
- IX.3 Chat (create a PDF by chat) — describe it, get a document
- IX.4 Chapter lab: from idea to illustrated PDF with TOC
- IX.5 How the family works together
Part X — Understand with AI
10-ai-understand/ · family overview
AI apps read your content and produce new content. Treat results as drafts. See II.7.
- X.1 Ask PDF / Chat with PDF — grounded Q&A over one document
- X.2 PDF Knowledge Base — question many documents at once
- X.3 Abstracts Generator — Academic / Technical / Business / Informative / Descriptive
- X.4 Cheat Sheet — one-page distillation
- X.5 Checklist — Tasks / Learning / Preparation / Analysis × four depth levels
- X.6 Writing good prompts and options for AI apps
- X.7 Trust, verification and limits of AI output
- X.8 Chapter lab: absorb a 300-page report in 30 minutes
Part XI — AI for Developers and Teams
11-ai-dev/ · family overview
For readers who work with code, repositories and releases.
- XI.1 Assembly Chat — ask questions of a .NET assembly's public API
- XI.2 Code Graph — see how a repository fits together
- XI.3 Repo Docs — generate documentation from a repository
- XI.4 Release Notes — retired; what to use instead
- XI.5 Refactor Assistant
- XI.6 NuGet Updater
- XI.7 Fork Sync — retired; what to use instead
- XI.8 Page SEO Checker
- XI.9 Advisor · Document Engineer (and retired Personal Engineer)
Part XII — Recipes: Reaching Goals with Chains of Apps
12-recipes/ · how to read a recipe
Every recipe uses one template: Goal → Starting state → Chain (diagram) → Step-by-step with options → Expected result → Check → Variations → What can go wrong.
XII.A Everyday office
- Send a big report by email — Delete Page → Optimize → Compressor
- Assemble one submission pack from many files — Converter → Rotation → Resize → Merger → Add TOC → Compressor
- Make a form-ready, fill-and-return PDF safe — Extract Form Data → Flatten → Protector
- Share only part of a document — Splitter / Delete Page → Protector
XII.B Scans and archives
- Phone photos to a clean searchable PDF — Converter (images→PDF) → Deskew → Rotation → Searchable → Compressor
- Digitise a paper archive — Deskew → Searchable → Extract Metadata → PDF/A Validator → PDF KB
- Recover and re-use content from a damaged file — Recovery → Unlock → Extract Text / Extract Image
XII.C Contracts, legal and compliance
- Contract round-trip — Extract Text diff → Flatten → Sign → Verify Signature
- Archive-ready (PDF/A) delivery — Flatten → Optimize → PDF/A Validator
- Safe external sharing checklist — Extract Metadata (inspect) → Delete Page → Recovery (Full sanitize) → Protector
XII.D Publishing and print
- Print a folded booklet — Resize → Booklet
- Turn a manuscript into an illustrated e-book — Chat/Converter → TOC AI → Illustrator → Add TOC → Compressor
- Create a poster/collage sheet — Extract Image → Collage → Resize
XII.E Knowledge work
- Understand a long document fast — Searchable (if scanned) → Abstracts → Cheat Sheet → Ask PDF
- Turn a spec into an action plan — Ask PDF → Checklist
- Screen candidates — Resume (score each CV) → compare scores
- Build a team knowledge base — Searchable → Merger → PDF KB
XII.F Data extraction
- Tables and forms to spreadsheet — Parser / Extract Form Data → Converter
- Harvest all images and text from a brochure
XII.G Designing your own chain
- Recipe-building worksheet: define goal, check state, pick apps, order, verify
- Repeat the chain every day: manual → CLI → MCP/API
Part XIII — Beyond the Browser
13-beyond-browser/
- XIII.1 Which channel when: Web app vs CLI vs MCP vs API vs Self-host
- XIII.2 Documentize CLI plugins
- Install, one command per app, chaining with shell pipes and scripts
- XIII.3 MCP server: use Documentize inside Claude, Cursor, VS Code
- Configure a client, the tool list, example conversations that run a chain
- XIII.4 REST API
- Keys, uploading, asynchronous tasks, status polling, downloading results, continuations
- XIII.5 Self-Host — same apps on your own server for private data
- XIII.6 Desktop
- XIII.7 Documentize.Plugins for developers
- XIII.8 The how-to guides library — same operation on web, CLI and MCP side by side
Part XIV — When Things Go Wrong
14-troubleshooting/
- XIV.1 Diagnosis flowchart: find the symptom, find the page
- XIV.2 Upload problems — size, type, password, corrupted
- XIV.3 "The result looks wrong" — fonts, missing text, lost bookmarks, broken forms
- XIV.4 "It stopped half way" — time limit, daily allowance, continuations
- XIV.5 Signatures, passwords and permissions problems
- XIV.6 OCR and AI quality problems
- XIV.7 Getting help: what to include in a support request
Part XV — Reference & Appendices
15-appendices/
- A. Format matrix: app × input/output
- B. Option dictionary: every option name, meaning and typical values
- C. App cheat sheets — one printable card per app
- D. Dependency matrix (full)
- E. Glossary — OCR, PDF/A, text layer, flatten, redact, outline, deskew, DPI, RAG, MCP…
- F. Limits and plans table
- G. FAQ
- H. Terms, licensing and acknowledgements
- I. What's new / roadmap
- J. Index of apps (A–Z) and index of goals
Template for every app page (Parts IV–XI)
Use identical headings so readers learn the pattern once.
- In one sentence — what it does
- What makes it unique — what no other Documentize app does
- When to use it / when not to — and the better app in that case
- Before you start — required document state (scanned? locked? size?)
- Step by step — with screenshot placeholders
- Options explained — each option, effect, default, recommended value
- What you get — output format, naming, what changed vs. original
- Quality & limits — lossless / lossy / AI-generated; size and time limits
- Works well with — use before / use after (links to neighbours and to recipes)
- Common mistakes & fixes
- Same thing elsewhere — CLI command, MCP tool, API call
- Try it — 5-minute exercise with a sample file
Authoring notes
- Names, options and limits must be checked against the live apps before each chapter is finalised.
- Keep Part III (the map) and Appendix D in sync — they are the book's backbone.
- Every app page links up to its family overview and across to at least two neighbours and one recipe.