How fonts work in PDFs

Fonts are one of the most common causes of PDF headaches. Text can shift, letters can look wrong, or characters can vanish entirely — almost always because of how (or whether) a font was embedded.

Embedded vs referenced fonts

A PDF can either embed a font — packaging the actual typeface data inside the file — or merely reference it by name and hope the reader's device has it. Embedded fonts render identically everywhere. Referenced fonts get substituted if they're missing, which is why a document can suddenly look different on another computer.

Font subsetting

Embedding an entire font can be large, so most tools subset: they include only the specific glyphs the document actually uses. A subsetted font keeps the file small while still displaying perfectly. The trade-off is that heavily subsetted PDFs are harder to edit, because the letters you'd want to type may not be present in the embedded subset.

The 14 standard fonts

The PDF spec defines 14 "standard" fonts (variants of Helvetica, Times, Courier, plus Symbol and ZapfDingbats) that every reader is expected to provide. Documents using only these don't need to embed anything — which is why they're a safe default when you add text to a PDF.

Common font problems

• Missing glyphs — a symbol shows as a box because it isn't in the subset.
• Substitution — a non-embedded font is swapped for a look-alike, changing spacing.
• Copy-paste garbage — text copies as gibberish when the font lacks proper character mapping.

When you need editable, reliable text, adding a fresh text layer in a standard font is often cleaner than trying to match a subsetted original.