How fancy text actually works
These are not fonts. A font is a rendering instruction that lives with the document and disappears when you paste elsewhere. What this tool produces is a substitution: each letter is swapped for a different character that already exists in Unicode and happens to look like a styled version of it. Bold A is a distinct codepoint from ordinary A, sitting in a block called Mathematical Alphanumeric Symbols that was added for typesetting equations.
That is why the styling survives. When you paste into a social bio, a chat message, a document title, or a form field, you are pasting the characters themselves, and any system that can display them will. No formatting is being applied, so nothing gets stripped by a plain-text field.
It is also why the results have gaps. Unicode never intended these blocks as a general styling mechanism, so the coverage is uneven. Only some styles include digits, most cover the Latin alphabet and nothing else, and a handful of letters simply do not exist in certain styles because nobody needed them for mathematics.
- Each style is a character substitution, not a font
- Styled characters survive a paste into plain-text fields
- Coverage is uneven because these blocks were made for mathematics
What each style covers
Seven of the fourteen come from the mathematical alphanumeric blocks: bold, italic, bold italic, script, fraktur, double-struck, and monospace. Circled and squared come from the enclosed alphanumeric blocks, and fullwidth comes from the block designed for mixing Latin letters with East Asian text at matching widths. Small caps and superscript are assembled from scattered individual characters rather than a contiguous block.
Digits are the biggest gap. Only bold, double-struck, monospace, fullwidth, and circled include styled numerals. In every other style, numbers pass through unchanged, so a bio containing a year will show that year in ordinary digits next to styled letters. That is a Unicode limitation, not something a different tool would solve.
The styles built from letter substitution only transform A to Z and a to z. Accented Latin letters, Cyrillic, Greek, punctuation, and emoji pass through untouched. The two exceptions are strikethrough and underline, which work by adding a combining mark after each character and therefore apply to almost anything, including non-Latin scripts.
A few individual quirks are worth knowing. Squared text maps lowercase to the same glyphs as uppercase, so casing is lost. Small caps and superscript are missing a few letters entirely because Unicode never assigned them. Several styles borrow older standalone characters for letters that were already encoded before the block existed, which is why script capital B and fraktur H look slightly different from their neighbors.
Where to use it, and where not to
The natural home for this is anywhere that gives you a single unformatted text field and no styling controls: social profile names and bios, chat handles, forum signatures, spreadsheet headers, and the occasional document heading where you want emphasis without touching the styles.
There is one important accessibility cost, and it is worth being straight about. Screen readers do not know that a bold mathematical A is meant to be the letter A. Depending on the reader, styled text is either spelled out character by character, read as a string of Unicode descriptions, or skipped entirely. A name in fancy text can become unreadable to a blind user.
The practical rule that follows: use it for decoration, never for content that carries meaning. A styled display name is fine. A styled paragraph, a styled call to action, or styled instructions are not. Some platforms also filter or normalize these characters in search indexes, so a fancy name can become harder to find.
- Best for bios, handles, signatures, and short decorative labels
- Screen readers often spell styled text out character by character
- Some platforms normalize or exclude these characters from search