How to use the Word Counter for Social Media
- Paste your text. Paste a caption, a thread, a LinkedIn draft or a newsletter intro. Paragraph breaks are kept and counted.
- Read the counts. Words, unique words, sentences, paragraphs and average words per sentence update as you type. Reading time at 200 and 250 words a minute and speaking time at 130 are shown as estimates.
- Check the fit. See roughly how many words of your style fit in X, Threads, Instagram and LinkedIn posts, based on the average word length in your own text.
- Trim to a target. Enter a word target and the helper cuts at the last full sentence that fits, so you keep clean endings. Copy the result.
How the word counter works
What it measures
Words are counted from whitespace-separated pieces of text that contain at least one letter or number, using Unicode letter and number classes, so accented and non-Latin alphabets count. Sentences end at a full stop, question mark, exclamation mark, ellipsis or CJK full stop, with a short exception list for decimals, initials and common abbreviations such as e.g. and Dr. A blank line also ends a sentence, so a headline without a full stop counts as its own sentence.
The top 10 words skip common English stop words, numbers, links, hashtags and mentions, and are ranked by count and then alphabetically. Hashtags and mentions are counted separately: a hashtag needs at least one letter, so #1 is not counted.
Limits and estimates
Network limits are 280 characters on X per X's character counting documentation, 500 on Threads per Meta's Threads developer documentation, and 2,200 for an Instagram caption per Meta's Instagram developer documentation, all checked October 2026. The LinkedIn figure of 3,000 is widely cited but approximate, because we could not confirm it on an official page.
Word fit divides each limit by your average word length plus one space. It ignores links, emoji and line breaks, so treat it as a planning number.
Limitations
The tool is tuned for English and other space-separated languages. Sentence detection can miss unusual abbreviations. Reading and speaking times are common estimates, not measurements. Your text stays in your browser and is never uploaded.
Frequently asked questions
How is this different from the character counter?
The Social Media Character Counter measures characters against each network's limit, which is what decides whether a post can be published. This tool counts words, which is what readers and speakers feel: reading time, sentence length, repeated words. It also estimates how many words fit in a network at your own average word length.
How are words counted?
A word is any group of characters between spaces that contains a letter or a number. So "don't" is one word, "well-known" is one word, "3.5" is one word, and a link counts as one word. A lone emoji or a dash is not a word, but a word with an emoji attached to it is.
Why is the reading time given as 200 and 250 words a minute?
Common estimates for silent reading of adult prose fall in roughly that range, and the true speed depends on the reader and the text. Speaking time uses 130 words a minute, a typical pace for a calm talk. All three are rules of thumb. Use them to compare drafts and to plan a script, not as an exact length.
Does it work for Chinese, Japanese or Korean?
Only partly. These scripts are normally written without spaces between words, so a space-based count undercounts. When the tool sees CJK characters it shows how many there are, and for those languages the character count is the more useful number.
How reliable is the "words that fit" number?
It is an estimate: the network limit divided by your average word length plus one space. A text of short words fits more words than a text of long words. X counts some characters, such as emoji and links, differently, and the LinkedIn figure is a widely cited 3,000 characters that is approximate. Use the Character Counter for the exact check before you publish.
How does the trim helper pick where to cut?
It keeps whole sentences, in order, until the next sentence would go over your word target, and drops the rest. If the first sentence alone is longer than the target, it cuts at the word instead and tells you it did.