Extract hashtags from any text, in any language
This hashtag extractor finds every hashtag in a caption, a batch of posts, a comment thread or a chat export and gives you a tidy list with a count for each tag. It is built for real social text: hashtags in Hindi, Tamil, Japanese, Arabic or accented Latin (#café, #दिवाली, #日本) are read whole, including the vowel signs that Indian scripts depend on, and the full-width ＃ used on Japanese and Chinese keyboards works too.
It also knows what is not a hashtag. The #section at the end of a web address, HTML character references such as ', numbers like #1 or #2024, words with a # glued into the middle and (optionally) colour codes such as #ff8800 are all left out. Merge #Travel with #travel, sort by popularity, then copy the set straight into your next post or download a CSV.
Nothing is uploaded — the extraction runs in your browser, so drafts and private messages stay with you.
How to use the Hashtag Extractor
- Paste your posts or captions into “Text containing hashtags”, or press “Open a file” for a .txt, .csv, .json or .html export.
- Keep “Merge capitalisation” ticked to count #Travel and #travel together, or untick it to see each spelling separately.
- Choose the order — most used, as found or A → Z — and whether to write the tags with or without the # sign.
- Pick a separator: spaces are ready to paste under a post; new lines or commas suit spreadsheets and notes.
- Copy the list, or download a CSV with each hashtag, its number of uses, its share of all tags and the first line it appears on.
Hashtag Extractor features
Every script, not just English
Letters from any alphabet plus combining marks are part of the tag, so #हिंदी and #தமிழ் are not cut short, and joiners used in Indic and Persian text are kept.
Smart exclusions
Ignores web-address fragments, HTML entities, email addresses, all-number tags (#1, #2024) and # characters stuck to the end of another word.
Counts and shares
A table shows how many times each tag appears and its share of all hashtags, with bars for a quick visual check.
Case and Unicode merging
#Mumbai, #mumbai and #MUMBAI merge into one entry shown in the most common spelling; precomposed and decomposed accents count as the same tag.
Paste-ready output
With or without #, separated by spaces, new lines or commas, in order of use, first appearance or alphabet.
What counts as a hashtag
The rules follow the way the big platforms link hashtags. A tag starts with # (or the full-width ＃) that is not stuck to a letter or digit before it, continues through letters, digits, underscores and combining marks, and stops at a space, punctuation or an emoji.
| In your text | Result | Why |
|---|---|---|
#SundayFunday! | #SundayFunday | Punctuation ends the tag |
#mom’s | #mom | An apostrophe ends the tag |
#travel_2024 | #travel_2024 | Underscores and digits are allowed |
#2024, #1 | not a hashtag | A tag needs at least one letter |
#दिवाली | #दिवाली | Vowel signs are part of the word |
example.com/page#top | not a hashtag | Part of a web address |
', #tag | not a hashtag | HTML character references |
#ff8800 | skipped by default | Looks like a colour code (hex digits with a number) |
Tips for better hashtags
- Write multi-word tags in CamelCase — #ThrowbackThursday rather than #throwbackthursday. Platforms treat both the same, but screen readers can pronounce the words separately.
- Check the count column before reposting: a tag you used ten times in a week may be worth replacing with a more specific one.
- Short, all-letter colour names such as #fff or #bad are kept, because they could be genuine tags; only hex codes that contain a digit are skipped.
- Collect the @handles from the same posts with the Mention Extractor, and tidy the caption itself with the Instagram Caption Formatter.
- Looking at the ordinary words instead? The Word Frequency Counter ranks them. Links and addresses in the same text are handled by the URL Extractor and Email Extractor.
Hashtag Extractor FAQ
How do I extract hashtags from a caption or post?
Paste the text into the hashtag extractor. Every hashtag appears straight away in “Extracted hashtags”, once each and most used first, ready to copy. The table underneath shows how often each one was used.
Does it work with Hindi, Tamil, Arabic or Japanese hashtags?
Yes. Any alphabet is supported, and the combining vowel signs used in Devanagari, Tamil and other Indian scripts stay inside the tag, so #हिंदी is extracted whole rather than cut after the first letter.
Are #Travel and #travel counted as the same hashtag?
With “Merge capitalisation” ticked (the default) they are counted together and shown in whichever spelling appears most often. Untick it to keep every spelling separate.
Why is #2024 not in my list?
Hashtags made only of digits are not treated as tags on the major platforms, so the extractor ignores them. Add a letter — #2024goals — and it is found.
Can I get the hashtags without the # sign?
Yes. Set “Write hashtags” to “Without #”. The CSV always includes the # so the tags stay recognisable.
Is my text sent anywhere?
No. The hashtag extractor works entirely in your browser; nothing is uploaded or stored.