Word Counter
Paste or type your text and see the word count update instantly, together with characters, sentences, paragraphs and estimated reading time.
Character limits by platform
| Platform | Limit | Remaining |
|---|---|---|
| X (Twitter) post | 280 | 280 |
| Instagram caption | 2,200 | 2,200 |
| Instagram bio | 150 | 150 |
| TikTok caption | 4,000 | 4,000 |
| LinkedIn post | 3,000 | 3,000 |
| YouTube title | 100 | 100 |
| Google title tag | 60 | 60 |
| Google meta description | 160 | 160 |
| SMS | 160 | 160 |
Character frequency
| Character | Count | % |
|---|---|---|
| — | ||
Keyword density
| Word | Count | % |
|---|---|---|
| — | ||
Useful for essays with a word limit, blog posts, meta descriptions and social media. The counter runs entirely in your browser, so your text is never uploaded, and the tool also works offline once the page is loaded.
What counts as a word here
The counter uses your browser’s built-in Unicode segmentation, the same machinery a browser uses to select a word when you double-click. Accents, non-Latin scripts and numbers are handled properly, and punctuation standing on its own is not counted.
One difference is worth knowing about: a hyphenated form such as "well-known" counts as two words here and as one in Microsoft Word. Neither is wrong, they are different conventions, and for most text the gap is under a percent. If you are within a few words of a hard limit, count in the program your reader will use.
The limits people are usually counting against
Academic abstracts typically run to 150 to 250 words, and journals enforce that strictly. Blog posts written for search commonly land between 1,000 and 2,000. University essays state their own limit and often allow 10% either side, though that allowance is a departmental rule rather than a universal one.
Where a limit is set in characters rather than words, and a search snippet, a social bio or an SMS all are, the character counter is the tool for it: it shows the remaining allowance against each platform’s limit as you type.
Frequently asked questions
How are words counted?
Words are detected with the Unicode segmentation rules built into your browser, so accented letters, numbers and non-Latin alphabets are handled correctly. Standalone punctuation is not counted as a word. Note that a hyphenated form like "well-known" counts as two words here, while Word or Google Docs count it as one, so totals can differ slightly.
Is my text uploaded anywhere?
No. Counting happens with JavaScript on your device. The text never leaves the page, and nothing is stored on our servers.
How is reading time estimated?
Reading time assumes an average pace of about 200 words per minute, rounded up to the next minute. Actual pace varies by reader and by how technical the text is.
Why does my word processor show a different total?
Because each program draws the line in its own place. Hyphenated forms, numbers with units, footnotes, headings, captions and text in tables are all treated differently from one tool to the next, and a difference of a percent or two on a long document is normal. When a limit is strict, count with the tool whoever set the limit will use.
Does my text survive a page reload?
Yes, within the same tab. What you type is kept in your browser’s session storage so that an accidental refresh does not lose it, and it is cleared when you close the tab. It is never transmitted, so it is not recoverable from another device or another browser.
What is the word frequency table for?
It lists the ten most common words in your text with their share of the total, which is the figure sometimes called keyword density. It is a quick way to catch a word you have leaned on too heavily, and a rough check when you are writing to a brief that asks for a term to appear.
Does it work for Chinese, Japanese or Korean?
It counts them, using the same Unicode segmentation your browser uses, which splits CJK text into word-like units rather than individual characters. If your limit is expressed in characters, which is the more usual convention for those languages, read the character figure instead of the word figure.