All guides

PDF & DocumentsUpdated 17 October 2026

PDFs and Word documents

Why PDFs refuse to shrink, what merging and redaction really do, reading a DOCX without Word, and what survives a conversion between them.

Nearly every complaint about PDFs traces back to one design decision: the format records where ink goes on a page, not what the page means. There are no paragraphs in a PDF, no headings and no tables, only positioned glyphs and images that a reader's eye assembles into those things. Text extraction returns odd line breaks because the line breaks were never recorded, a scan has no text at all until OCR puts some there, and compression barely touches a text-only document while transforming an image-heavy one. A Word file is the opposite: a .docx is a ZIP of XML that records structure and leaves the final layout to whichever program opens it. That is why converting Word to PDF fixes how a document looks, and why converting back has to guess the structure. These guides work through both formats and the hops between them.

15 guides5 subjects

15 guides

Taking pages out of a PDF: split, delete, extract, and which you want

Three buttons that look like the same operation. The difference is how many files come out and what happens to everything you did not select.

4 min

Page numbers, headers and margins: making a PDF look submitted, not exported

A PDF page has four different boundaries and a rotation value that does not move any content. Both explain why stamped text lands in odd places.

4 min

Making a PDF out of HTML, plain text and a form you filled in

A web page has no pages. Turning one into a PDF means inventing page breaks, and where they land is the whole problem.

4 min

Turning photos into a PDF, and finding what changed between two versions

Twelve photos of pages is not a document until something gives them an order, a page size and a sensible resolution.

4 min

Getting a table out of a PDF without retyping it

A table in a PDF is not a table. It is text at coordinates that happens to line up, which is why copying it gives you one long column.

4 min

Word to PDF and back: what survives the round trip

A PDF made from a Word file has no paragraphs, no styles and no reading order, only glyphs at coordinates. That single fact explains every frustration with converting one back.

5 min

Opening a DOCX without Word, and what is hiding inside it

Rename a .docx to .zip and open it. Everything Word knows about your document is sitting there in plain XML, including who wrote it and how long they took.

4 min

Writing in Markdown and generating the Word file later

Word documents cannot be diffed, merged or reviewed line by line. Plain text can. The trick is generating the DOCX at the end rather than living in it.

4 min

Word styles: why formatting by hand costs you later

Bold 16pt text looks like a heading and is not one. Nothing downstream can tell the difference, which is where the trouble starts.

4 min

Finding what changed between two versions of a document

A comparison that reports the whole document changed usually means one paragraph moved and everything after it reflowed.

4 min

How PDF compression actually works (and why some files refuse to shrink)

Every "compress PDF" button does roughly four things. Understanding them tells you in advance whether your 40 MB file will drop to 3 MB or stubbornly stay at 39 MB.

3 min

Merging PDFs: what gets preserved, what quietly breaks

Merging looks trivial until a signed form comes out blank or a 3 MB and 4 MB file produce a 12 MB result. Both have the same cause.

3 min

How to redact a PDF properly (a black rectangle is not redaction)

Courts, police departments and law firms have all published documents whose "redacted" passages could be read by selecting the text and pasting it elsewhere. The mistake is always the same.

3 min

OCR on scanned PDFs: what it can read, what it cannot, and why DPI matters

A scanned PDF is a stack of photographs. OCR is what turns those photographs back into words, when the input gives it a fair chance.

3 min

Getting a PDF under a strict upload limit, without destroying it

When a portal rejects your document at 1.2 MB against a 1 MB cap, panic-compressing to the lowest setting is the wrong move. There is a better order to try things in.

3 min