LinkedIn text formatter

There is no hidden formatting code in the text you copy. Each styled letter is a separate Unicode character that happens to be drawn bold or slanted. Unicode added these in the Mathematical Alphanumeric Symbols block (U+1D400 to U+1D7FF) so math notation could tell a bold variable from an italic one. The Unicode 18.0 code chart, read Nov 1, 2026, says it plainly: the block is "to be used for mathematical variables where style variations are important semantically. For general text, use standard Latin and Greek letters with markup."

Social posts have no markup, so people borrow the math letters anyway. Here is exactly what each button does:

• Bold uses Mathematical Sans-Serif Bold (capital A is U+1D5D4), plus sans-serif bold digits.
• Italic uses Mathematical Sans-Serif Italic (capital A is U+1D608). Letters only.
• Bold italic uses Mathematical Sans-Serif Bold Italic (capital A is U+1D63C). Letters only.
• Mono uses Mathematical Monospace (capital A is U+1D670), plus monospace digits.
• Underline and Strike keep your letters and add a combining mark after each character: U+0332 (combining low line) or U+0336 (combining long stroke overlay). Spaces and line breaks are skipped.
• Clear maps every styled character back to plain A to Z, a to z and 0 to 9 and removes the underline and strike marks, so you can switch styles without retyping.

Only the 26 basic Latin letters (both cases) and, for bold and mono, the digits 0 to 9 change. Everything else passes through untouched. Bold turns "Café déjà vu, 3 ideas" into 𝗖𝗮𝗳é 𝗱é𝗷à 𝘃𝘂, 𝟯 𝗶𝗱𝗲𝗮𝘀: the é and à stay regular weight in the middle of bold words.

• No italic digits. Unicode defines digit sets only for bold, double-struck, sans-serif, sans-serif bold and monospace. There are no sans-serif italic or bold italic digits, so "Hello 2026" in italic comes out as 𝘏𝘦𝘭𝘭𝘰 2026.
• No accented letters. The block has no styled é, ñ, ü or ø, so names and words in many languages end up half styled. If that looks wrong, pick a different word to emphasize.
• No punctuation, symbols or emoji. Colons, commas, # and @ stay as they are in bold, italic and mono. Underline and strike do mark punctuation and digits, since they add a mark rather than swap the letter.
• Greek has bold and italic forms in the same block, but this tool converts Latin letters and digits only.

This is the main reason to keep styled text short. Accessibility consultant Adrian Roselli tested fake bold and italic with default settings (his post was updated Sep 29, 2025; we read it Nov 1, 2026). NVDA with Firefox, VoiceOver on macOS and iPadOS, and Narrator with Edge did not announce the styled text. JAWS with Chrome announced each letter individually, with its case and style. Only TalkBack with Chrome read it as ordinary words.

Scope for Business, writing for the UK disability charity Scope, says these characters are "inaccessible to screen reader users" and gives the example of a letter announced as "mathematical sans-serif script t". It also notes fancy characters are harder to read for people with dyslexia or low vision. Its advice is to use standard text when you want a wide audience to understand the message.

So the safe pattern is decoration on words the post doesn't depend on. If a screen reader drops your bold label, the sentence after it should still make sense. If the bold words carry the whole point, write them in plain text.

A bold 𝘄𝗲𝗯𝗵𝗼𝗼𝗸𝘀 is a different string of characters from the word webhooks. The Unicode chart lists a compatibility mapping from each styled letter back to its plain letter, and software that applies that normalization (NFKC) can fold them together. Plenty of software doesn't. LinkedIn doesn't document how its search treats these characters, so don't count on styled words being found by people searching for them.

The practical rules follow from that. Leave keywords, product names and your own name in plain text. Don't select hashtags: the # stays a #, but the letters after it become different characters, so the tag no longer matches the plain one. The same goes for anything a reader might copy into another tool, such as a command, a URL or an email address.

A rule of thumb: one styled element per paragraph at most, and never a whole paragraph.

• Bold for a short label at the start of a line ("What changed:", "Key takeaway:") or a single hook line at the top of the post.
• Italic for one word of stress, a title, or a short aside. Remember it skips digits.
• Bold italic rarely. It is heavier to read than bold and adds nothing bold can't do.
• Mono for short technical strings that should look like code, such as an endpoint name. Keep the real, copyable version in plain text if readers need to use it.
• Strikethrough for a visible correction, like an old price or version number next to the new one.
• Underline almost never. Readers expect underlined text to be a link, and the combining marks render unevenly across fonts.

These were generated with the buttons above, so they match what the tool produces. A bolded label followed by plain text keeps the sentence readable when the label is skipped. If you're unsure where LinkedIn will cut the post, paste the result into the LinkedIn post preview.

𝗪𝗵𝗮𝘁 𝗰𝗵𝗮𝗻𝗴𝗲𝗱: failed posts now retry five times before we mark them failed.

𝗞𝗲𝘆 𝘁𝗮𝗸𝗲𝗮𝘄𝗮𝘆: check the webhook, not the dashboard.

The endpoint is 𝙿𝙾𝚂𝚃 /𝚟𝟷/𝚙𝚘𝚜𝚝𝚜 (plain: POST /v1/posts).

This is 𝘯𝘰𝘵 a scheduling app.

𝗕𝗲𝗳𝗼𝗿𝗲: v̶2̶  𝗔𝗳𝘁𝗲𝗿: v3

The formatter keeps your line breaks and spaces exactly as typed, and underline and strike skip them, so paragraph spacing survives the copy. Blank lines between short paragraphs make a post easier to scan than any amount of bold. Check the pasted result in LinkedIn's composer before you publish, since that is the rendering readers see.

LinkedIn Help (read Nov 1, 2026) says posts can be up to 3,000 characters. The counter under the tool counts graphemes, the characters you see, which is also how our API counts LinkedIn posts. Styled letters are heavier underneath: "Hello 2026" in bold is 10 graphemes but 19 UTF-16 code units and 37 UTF-8 bytes. LinkedIn doesn't say which unit its limit uses, so leave some room if a heavily styled post sits near 3,000. More on the limit in our LinkedIn character limit guide.

The LinkedIn Posts API takes post text in its "little" text format. Microsoft's documentation for it (read Nov 1, 2026) lists four elements: plain text, mentions, hashtags and a hashtag template. There is no bold or italic element, so Unicode styling is also the only way to emphasize words in a post created through the API.

Styled letters are ordinary text to the API and pass through unchanged. What does need care is the reserved character set: little text requires characters like ( ) [ ] { } < > @ * _ ~ | and \ to be escaped with a backslash, even when they aren't part of a mention. Our LinkedIn API integration escapes them for you and leaves hashtags intact. For the text below, the commentary we send is:

// you send: "𝗪𝗵𝗮𝘁 𝗰𝗵𝗮𝗻𝗴𝗲𝗱: failed posts retry (5 times) #api"
{ "commentary": "𝗪𝗵𝗮𝘁 𝗰𝗵𝗮𝗻𝗴𝗲𝗱: failed posts retry \\(5 times\\) #api" }