Why can’t I edit this PDF?

1. It’s a scan, not a document

By far the most common answer. Somebody put paper through a scanner or photographed it with a phone, and what you have is a picture of a page wrapped in a PDF. There is no text in the file at all, only pixels that look like text to you and mean nothing to the computer.

How to tell: you cannot select any text, the whole page selects as one object, and zooming in a long way makes the letters go soft and blocky rather than staying crisp.

What to do: run OCR — optical character recognition — which looks at the picture and works out what the letters were, then writes a text layer behind the image. Afterwards you can search and copy. Editing is a different matter: OCR text sits invisibly behind a picture of the original, so changing a word leaves the old word still showing in the image underneath. Genuinely editing a scan means rebuilding the page, and you should expect it to look different.

Free OCR worth knowing about: Tesseract and the tools built on it, OCRmyPDF, and the OCR built into Microsoft OneNote and into most scanner software. Emifyn PDF Studio does not currently do OCR — it is on the list, and saying so is more use to you than pretending otherwise.

2. Your program can only draw on top

Most free PDF software is a viewer that has learned to annotate. It can add a highlight, a sticky note, a signature and a new text box, and it cannot touch a word that was already on the page. Nothing is wrong with your file; you are simply asking the wrong program.

How to tell: clicking a paragraph does nothing, or draws a new empty box beside it rather than putting a cursor into the existing words.

What to do: use something that rewrites the page’s own content. Editing the text in a PDF covers the free options on Windows.

3. The document has a permissions password

A PDF can carry restrictions that say no printing, no copying, no editing. The file opens perfectly normally — you are not asked for anything — but well-behaved software greys out the tools.

How to tell: the file opens straight away, yet the edit tools are disabled. Most readers will show it somewhere under document properties or security, often as a padlock in the title bar or status bar.

What to do: if it is your document, or you have the owner password, you can lift the restrictions. Removing a password from a PDF you own explains the two different kinds and which is which. If it is not your document, the restriction is telling you something and the right move is to ask whoever sent it.

4. The font isn’t embedded, or is missing the letters you need

This one is invisible and catches people out constantly.

A PDF is supposed to carry its fonts inside it so the document looks the same everywhere. To keep the file small, it usually carries a subset — only the characters the document actually used. If a document contains no ampersand, its embedded font contains no ampersand, and typing one is not a matter of pressing the key. The editor has to open the embedded font programme, add the glyph, and write it back.

Sometimes the font is not embedded at all, only named, and the file is relying on your computer having it. If you do not, your reader substitutes something approximate and the page you are looking at is already not quite the page the author made.

How to tell: the text selects normally but the editor refuses the paragraph, complains about a font, or replaces your new characters with blanks or boxes. Document properties will usually list the fonts and say “Embedded Subset”, “Embedded”, or nothing.

What to do: use an editor that handles subset fonts properly rather than one that gives up or guesses. Emifyn extends the embedded font where it can, falls back to the same font installed on your PC, and — if it has to use something else — says which. If nothing will do, retype the block in a font you do have and accept that it will look slightly different.

5. The text was converted to outlines

Design and print workflows often “convert text to outlines” or “flatten transparency” on export, which turns every letter into a vector shape. It guarantees the printer sees exactly what the designer saw. It also means the document no longer contains any text, only drawings that happen to be letter-shaped.

How to tell: the text looks perfectly sharp at any zoom — so it is not a scan — but nothing selects, and searching for a word you can plainly see finds no results.

What to do: there is no recovering it. You can OCR it like a scan, or you can cover the old text and set new text on top. If it is your own artwork, go back to the source file.

6. It’s built in an unusual way

A handful of documents are structurally awkward:

• Type 3 fonts, where each character is its own miniature drawing routine rather than a normal glyph outline. Old, rare, and hard to extend.
• Text split across many tiny positioned runs, common in output from older typesetting systems, where a single visible sentence is thirty separate drawing commands and the word boundaries have to be inferred.
• XFA forms — a form technology Adobe layered on top of PDF. Non-Adobe software often shows only a “please open this in Adobe Reader” message.
• Damaged files, where the internal index no longer matches the contents. Many tools can repair these; it is worth trying before assuming the worst.

What to do: for the first two, a good editor will often manage and a poor one will not. For XFA, Adobe’s own reader is genuinely the only reliable answer. For damage, run a repair and try again.

A shortcut for checking

Emifyn PDF Studio has a key for exactly this question. Ctrl+B draws a box round every paragraph on the page: normal boxes are text it can rewrite, red boxes are text it would have to refuse. It is a ten-second way of finding out which of the six situations above you are in, before you have invested any work in the document.