Guides: how PDF actually works
Most "how to" articles about PDF tell you which button to press. These do something different: they explain what the file does underneath, so the button makes sense — why a merged PDF is not a concatenation, what a password does and does not protect, why compression sometimes changes nothing, and why a black rectangle is not redaction.
Every claim with a number in it was measured with the tools on this site, on files you could reproduce, and the guides say when the honest answer is "it depends" or "use something else". They are written and maintained by the ZasPDF team; the about page explains who that is and how the articles are checked. Who writes these guides →
Fundamentals
The structure of a PDF and the consequences of that structure for everyday tasks.
What PDF/A is, why your "normal" PDF probably isn't one, and how to check
PDF/A is the archival flavour of PDF that courts, registries and universities ask for. It forbids most of what makes an ordinary PDF convenient. Here is what it requires, why a file exported from Word usually fails, and how to verify one without paying for Acrobat.
"Print to PDF" and "Export as PDF" produce different files. Here is what changes
Both buttons give you a PDF that looks the same on screen. Underneath, one keeps text, links, bookmarks and structure; the other keeps only what a printer would see. This guide explains the mechanism, how to tell which one you have, and when each is the right choice.
Size and quality
What compression, image sampling and font embedding actually do to a file.
Pages and structure
Merging, splitting, rotation, page order and what each of them changes inside the file.
Why merging two PDFs is harder than it looks (and how to do it losslessly)
A PDF is not a stream of pages you can staple together. Here is what a merge tool actually has to do, why naive concatenation fails, and how to combine files without losing a single pixel.
Splitting a PDF: extraction, not division, and why the distinction matters
Two problems hide behind the word "split": pulling out a page range, and separating a bulk scan into the documents it contains. They are solved differently, and only one of them a tool can do for you.
Why a PDF page "rotates by itself" on some devices, and what rotating actually changes
A page that looks upright in one viewer and sideways in another is not corrupted. PDF stores orientation as a separate instruction from the page content, and viewers, printers and phones apply it differently. This guide explains the /Rotate entry, the difference between rotating a page and rotating its content, and how to fix a document for good.
Conversions
Word, Excel, PowerPoint, images and OCR: what can be recovered from a PDF and what cannot.
Why OCR fails on skewed, dark or low-resolution scans (and what to fix first)
OCR engines are not bad at reading; they are bad at reading what a phone camera gives them. This guide explains the four things that break recognition — resolution, skew, contrast and margins — with the numbers behind each, and the order in which to fix them.
Why converting Word to PDF changes your layout, and how to get the same page every time
Fonts get substituted, line breaks move, a two-page letter becomes three. None of it is random. This guide explains the four mechanisms that reshape a document on the way to PDF, which converters are affected, and the settings that make the output match the screen.
Why tables from a PDF break in Excel: there are no tables in a PDF
A PDF does not contain tables, rows or cells — only characters at coordinates. Every converter has to guess the grid back. This guide explains what a converter can see, why merged cells, wrapped text and scanned pages defeat it, and how to get a clean spreadsheet out anyway.
Privacy and security
Passwords, encryption, metadata, redaction and what leaves your device when you use a cloud tool.
What actually happens when you upload a document to a free PDF tool
Not a scare story — a walk through the concrete steps a document goes through once it leaves your device, what a privacy policy actually has to disclose, and the questions worth asking before you upload anything.
RC4, AES-128, AES-256: what PDF password protection actually does
PDF encryption has a history of getting weaker as it gets more convenient, and it has a much bigger asterisk than most people realise: two different passwords, only one of which the cipher actually protects.
We covered a Social Security number with a black box. The text was still there.
A concrete, reproducible test: draw an opaque rectangle over sensitive text in a PDF, then run ordinary text extraction on the result. What comes back is the reason redaction and covering are not the same operation.
A drawn signature and a digital signature are different things. Here is which one you need
A signature drawn on screen and a certificate-based digital signature look similar on the page and have almost nothing in common underneath. This guide explains what each one actually is, what the law generally says about them, and when a picture of your signature is enough.
The metadata hiding in your PDF: author names, software, dates — and how to see and remove it
A PDF carries more than its pages. The Info dictionary and the XMP block record who created it, with what, when, and sometimes the original filename or a previous title. This guide shows what is there, how to inspect it in any viewer, and the three ways to get rid of it.
Watermarks that actually do something, and the ones that come off in two clicks
A watermark is not encryption and not a signature; it is a mark on the page whose value depends entirely on what you expect it to do. This guide separates the three jobs a watermark can do — labelling, deterrence, tracing — and explains what makes a mark hard to remove and what makes it decorative.
What "unlocking" a PDF actually does: two passwords, one key, and the limits of the word
A PDF can have an open password, a permissions password, both or neither, and "unlock" means something different for each. This guide explains how the two passwords relate to the encryption key, why permission restrictions are so easy to strip, why an unknown open password is not, and where the ethical line sits.