What a PDF actually is (and why it behaves the way it does)
PDF is the most-shared document format in the world and the least understood. Knowing how it stores a page explains almost every frustration people have with it.
Last updated September 2026
Fundamentals · 6 min read
A PDF is a set of drawing instructions, not a document
A Word file describes content: this is a heading, this is a paragraph, reflow it to fit the window. A PDF describes appearance: place this glyph at this coordinate, in this font, at this size. The format was designed in 1993 so a page would print identically everywhere, and it achieves that by fixing everything in place.
That single design decision explains the format's reputation. Pages look identical on every device because nothing is allowed to move. Editing is awkward for exactly the same reason: there are no paragraphs to edit, only positioned glyphs. When an editor 'edits' PDF text, it is recalculating positions for you.
The three things inside a page
- Text objects — characters with a font reference and x/y coordinates. This is the 'text layer' and it is what makes a PDF searchable and selectable.
- Vector graphics — lines, curves and fills described mathematically. They stay crisp at any zoom and cost almost nothing in file size.
- Raster images — photos and scans stored as pixel data. These dominate file size and are the first thing to look at when a PDF is unexpectedly large.
Digital PDFs versus scanned PDFs
A digital PDF was exported from software, so it carries a real text layer. Try selecting a sentence: if the cursor highlights words, the text layer exists and conversion tools can read it accurately.
A scanned PDF is a photograph of paper wrapped in a PDF container. It looks like text to you and like a picture to software. Nothing can extract its words without optical character recognition (OCR), which guesses characters from pixel shapes. This is the single most common reason a PDF-to-Word or PDF-to-Text conversion comes back empty.
Fonts, and why text sometimes looks wrong
PDFs can embed the fonts they use, storing the actual glyph outlines in the file. Embedded fonts render identically anywhere and add to file size. When a font is not embedded, the reader substitutes something similar, which shifts spacing and line breaks. If a document must look exact — a contract, a certificate, a print job — embed the fonts when exporting.
What this means in practice
- If you can select the text, conversion tools will work well. If you cannot, you need OCR first.
- Large file? Look at images before anything else — they are almost always the cause.
- Need the layout preserved exactly? Keep it as a PDF and rearrange pages rather than converting to Word and back.
- Need to edit the words? Convert to Word, edit there, and export a fresh PDF.