Character Counter for Captions & Bios

Character limits are the one constraint in social writing that is completely non-negotiable, and the one you discover last - usually at the moment the composer refuses to accept the sentence you spent ten minutes on. Counting as you write is cheaper than rewriting after you are told.

This counter measures against a specific platform rather than in the abstract. Pick Instagram caption, X, TikTok caption, LinkedIn post, or Instagram bio, and the limit, the remaining allowance, and the warning colour all follow that choice. Alongside them you get characters with and without spaces, a word count, and a hashtag tally.

The counting is deliberately transparent, which also means it is deliberately naive in a couple of places. Knowing where it differs from the platform’s own count is the difference between a tool you can trust and one that surprises you at publish time.

The five presets are not interchangeable views of the same text - they represent genuinely different writing problems. At 2,200 characters an Instagram caption is effectively unconstrained, and the number that matters is not the cap but the roughly 125 characters visible before the "more" cut. At 280 on X every clause is a trade. At 150 in an Instagram bio you are writing a headline, not a paragraph.

So the useful reading depends on the preset. For the long caps, watch the remaining figure only to confirm you are nowhere near it, and spend your attention on the first line. For X and the bio, watch it constantly. The counter turns amber inside the last 40 characters and red once you are over, which is a reasonable prompt to start cutting rather than to start abbreviating.

One practical habit: draft in the preset with the tightest relevant limit. Text written to fit 280 characters can be expanded for LinkedIn far more easily than a 900-character LinkedIn post can be compressed to fit X, and the compression usually destroys the rhythm rather than the content.

The hashtag figure counts occurrences of a hash symbol followed by one or more word characters - letters, digits, or underscores. It is a count of tag-shaped tokens in your text, which is almost always what you want and is worth understanding at the edges. A hash followed by a space is not counted. A tag containing a hyphen is counted only up to the hyphen. Accented characters end the match, so a non-English tag can be undercounted.

Duplicates are counted separately, so pasting the same tag twice reports two. That is useful for catching accidental repetition in a long block, and it means the figure is not a count of distinct tags.

What the tally is for is spotting the two problems that a character count alone hides: a caption whose length is mostly tags rather than writing, and a tag block that has drifted past the platform’s allowance. It does not check whether the tags are relevant, restricted, or spelled correctly - those are separate questions with separate tools.

Characters with spaces is the raw string length, measured in UTF-16 code units. Characters without spaces removes every whitespace character - spaces, tabs, and newlines - and measures what is left. Words is the count of segments after trimming the text and splitting it on any run of whitespace, so an empty field reports zero and hyphenated compounds count as one word.

The limit is looked up from the platform preset you have selected, and remaining is that limit minus the with-spaces count. Unlike the bio preview, this figure is allowed to go negative, and the message below the grid switches from "characters left" to "characters over" when it does. The line colours amber below 40 remaining and red below zero.

Everything recomputes on each keystroke from the same string, locally, with no network request. The one thing worth flagging is that all of it operates on code units rather than on what a reader would call a character: a plain emoji measures two, a flag or skin-tone variant measures four or more, and a combining accent measures separately from the letter it sits on.