Almost every complaint about reading a PDF traces back to one design decision made in 1993. A PDF does not describe a document; it describes a printed page. Every character has a position measured from the corner of a sheet of a fixed size, and the format's entire promise is that the page will look identical everywhere.
That promise is why exam boards, journals and governments use it, and it is also why it reads badly on a phone. An ebook reflows its text to fit whatever screen it finds. A web page reshapes itself. A PDF cannot, because reflowing would break the guarantee it exists to make.
Once you accept that, the problem becomes tractable. You are not fighting a badly designed format; you are reading a description of a sheet of A4 on a device that is not a sheet of A4, and there are several good ways to close that gap.
Key facts
- What a PDF stores
- A fixed page layout, not flowing text
- Why text does not reflow
- Reflowing would break the identical-everywhere guarantee
- Best reading device
- A tablet of roughly 10 inches, with the margins cropped
- Worst reading device
- A phone, for anything laid out for A4
- Fastest comfort improvement
- Crop 15-20 mm off each margin
- When search will not work
- The document is a scan and has no text layer
The fixed-layout problem, stated properly
When software writes a PDF, it records where every glyph sits on a page of a stated size. There is no paragraph in the file, no sentence, often not even a word - just characters at coordinates, plus enough structure that a reader can usually work out the reading order afterwards.
This is why selecting text in a PDF sometimes produces something strange, why a two-column paper copies as interleaved nonsense, and why no reader can simply make the text bigger without also making the page bigger. The format was designed so that a page printed in Tokyo matches a page printed in Toronto, and it delivers that completely.
Reading a document laid out for A4
Effective comfort at typical viewing distances, assuming 11pt body text.
| Device | Whole page fits? | Comfortable? | What actually helps |
|---|---|---|---|
| Phone | Technically | No | Convert to text, or accept continuous zooming |
| 7-inch tablet | Yes | Marginal | Crop the margins; read in portrait |
| 10-inch tablet | Yes | Yes, once cropped | Crop, then nothing else |
| 13-inch laptop | Yes | Yes | Two-page view for anything you are comparing |
| E-ink reader | Yes | Yes, slowly | Crop hard; refresh rate punishes scrolling |
Cropping is the single best intervention
A document designed for A4 typically carries 20 to 25 millimetres of margin on every side. On a ten-inch tablet that margin occupies about a third of the glass and contains nothing.
Trimming it moves the text into the space the margin was wasting. The effective text size roughly doubles, and you have not zoomed, scrolled sideways or changed the document in any way that matters. Keep the cropped copy for reading and the original for printing.
Tools that help you read
Choosing a reader
Every operating system now ships a competent PDF reader, and for most documents the built-in one is the right answer. The cases where it is not are specific: very large documents, documents you need to compare side by side, and documents with a structure worth navigating.
The browser reader on this site exists for a narrower reason. It opens a file without importing it, without an account, and without the file leaving your machine, which makes it the fastest way to answer 'what is actually in this thing' before deciding what to do with it.
Built-in reader or a dedicated one
The honest split.
The one your device already has
- Opens instantly, no setup, no account
- Fine for reading, printing and filling simple forms
- Search works wherever a text layer exists
- Right answer for the large majority of documents
Something more capable
- Worth it for documents over a few hundred pages
- Side-by-side comparison of two documents
- Persistent bookmarks across a library
- Overkill for a folder of past papers
When search does not work
Searching a PDF finds characters in its text layer. A scanned document has no text layer: it is a photograph of a page, and to a computer it contains no letters at all, only pixels that a human happens to recognise.
This catches people out constantly, because a scan looks identical to a real document on screen. The test takes two seconds - try to select a word. If the cursor sweeps across as though over a picture, it is a scan, and no amount of searching will ever find anything in it.
The two-second test
Try to select a word with your cursor. If it highlights, the document has a text layer and can be searched, extracted and redacted. If it does not, it is an image of a page, and only optical character recognition would change that.
Reading with a screen reader
A well-made PDF carries a structure tree: headings marked as headings, tables marked as tables, images carrying alternative text. A screen reader uses that to move through the document sensibly, and the experience is genuinely good.
A badly made PDF carries none of it, and the same screen reader reads characters in whatever order they were written to the file - which on a two-column page means alternating between columns, sentence by sentence. This is not a screen reader limitation; it is a document that was never tagged. Where you have a choice, ask the source for a tagged PDF or an accessible alternative, because no reader can reconstruct structure that was never recorded.
Making a document comfortable to read
In the order worth trying them.
- Check whether it has a text layer - select a word
- Rotate any sideways pages and save the rotation into the file
- Crop 15 to 20 mm off each margin for tablet reading
- Use continuous scroll rather than page-by-page for prose, and the reverse for question papers
- For a phone, extract the text instead of fighting the layout
- Keep the original uncropped file for printing
Guides in this topic
Every one written by a named author and checked against real documents.
Choosing a PDF reader
For most documents, the one already on your device. The question is normally asked because a particular file is misbehaving, and the file is usually the problem.
How to tell whether a PDF is scanned
It takes two seconds, it determines what you can do with the file, and it explains most of the things people think are broken.
Searching inside a PDF
Search works on characters in the text layer. Everything that goes wrong with it comes from that sentence, including searching across a whole folder at once.
Reading a PDF on a phone
An A4 page on a phone screen is about a third of the size it was designed to be. There are three workable responses and one that wastes an hour.
Why PDF text does not reflow
It is not an oversight and it is not fixable. A PDF describes a printed page, and reflowing would break the only promise the format makes.
Questions and answers
Why can I not make the text bigger without zooming?
Because the text has no independent size. It is drawn at a fixed position on a fixed page, so making it bigger means making the page bigger, which is what zooming does. Cropping the margins achieves the same comfort without the side effects.
Why does copying from a two-column PDF produce jumbled text?
Because the file stores characters at positions, not paragraphs. A reader guesses the reading order, and on a two-column layout that guess often runs straight across both columns. Extracting with block layout preserved gets closer.
Is there a way to make a PDF reflow like an ebook?
Some readers attempt it on well-tagged documents, with mixed results, and most exam material is not tagged. Converting to plain text is the reliable route when layout genuinely does not matter, and accepting the fixed layout is the right answer when it does.
Why does my PDF look different on my phone than my laptop?
It should not, and if it does, the likely cause is a font that was not embedded, leaving the reader to substitute something similar. Everything else - spacing, line breaks, page boundaries - is fixed by the file.
Which is the best PDF reader?
The one already on your device, for almost everything. The question is usually asked when a specific document is misbehaving, and in that case the problem is generally the document rather than the reader.