Guide
Six TikTok captions from one line about your video
The caption under a TikTok and the words burned over it are two different jobs, and only one of them is written in the post box. This page writes the post box kind. Say what the video is about, pick a tone, and you get six captions in six shapes, each counted against the roughly one hundred characters that actually show before “more”.
TikTok caption writer
Runs in your browser · no account needed
Six captions, six different shapes, from one line about your video
Say what the video is about, add your ask if you have one, and pick a tone. You get a hook, a question, a statement with the ask attached, a list tease, a contrarian line and a plain descriptive one, each counted against TikTok’s limit and against the point where the feed hides the rest.
Two things are called captions
This writes the post text — the words in the feed, under or beside the video, that people read before deciding to watch.
It does not write the words burned over the footage. Those are subtitles, they follow your voice, and you can try five styles of them at the caption style preview.
- Characters TikTok's caption box accepts
- 2,200
- Characters that show before the fold, give or take
- 100
- Caption shapes the writer above returns
- 6
Two things are called captions, and you probably want this one
Search for a caption generator and you land on two completely different products wearing the same word.
One writes the post text: the line that sits in the feed, next to your username, under the video. That is this page.
The other burns subtitles into the frame, timed to your voice, one word or one line at a time. That is at /tools/caption-style-preview, where you can play your own sentence in five styles.
On TikTok the video is already playing by the time the post text is read. So it is not a headline waiting to be clicked. It runs alongside the video, explaining, asking or provoking while the thing plays.
That changes what a good one looks like. A headline announces. A caption keeps somebody in the room, or asks them for something on the way out.
The table sorts the two apart, so nobody spends an afternoon on the wrong one.
Post text and on-screen subtitles, side by side
Both are called captions. They live in different places, do different jobs and are made in different tools.
| The post text (this page) | The on-screen subtitles | |
|---|---|---|
| Where it lives | In the feed, over the bottom of the video | Burned into the frame, over the footage |
| What it is made of | Whatever you type into the post box | Your own words, timed to your own audio |
| How long it runs | 2,200 characters, roughly 100 of them visible | As long as the video, a word or a line at a time |
| Who it is written for | Somebody deciding whether to stay, and anybody searching later | Everybody watching with the sound off |
| Where you make it | Right here, then paste it into the post box | In an editor, or previewed at /tools/caption-style-preview |
The first line is the caption. The other 2,100 characters are a footnote.
TikTok's caption box takes 2,200 characters. It used to stop at 150.
That change did not make long captions good. It made the fold matter more.
The feed shows roughly the first line and then a “more” link. Tapping it is a deliberate act, and most people scrolling do not perform deliberate acts.
So the working ceiling is about a hundred characters. Everything past it is written for the small share of viewers who tap.
Exactly where the cut falls moves with the device, the font size and how long your username is, which is why every number here says “about”. The shape of the rule does not move: one line shows, the rest hides.
The writer above counts both numbers on every caption it gives you, and tells you plainly which one you have broken.
The two numbers a TikTok caption lives between
The limit is what the box accepts. The fold is what the feed shows. Only one of them changes what you should write.
| What TikTok accepts | What the feed shows | |
|---|---|---|
| Characters | 2,200 | roughly the first 100, then “more” |
| Hashtags | counted inside the same 2,200 | they push your words behind the fold |
| Line breaks | allowed, as many as you like | the second line is usually already hidden |
| Who reads past it | anybody who taps “more” | a small share of the people who watched |
| What that means for you | write as long as you want | write as though you have one line |
This is the whole editor
Highlight a phrase and a clip lands on those exact words. No timeline, no keyframes, no layers.
Six shapes, because a caption has a job
Most caption tools give you six rewrites of the same sentence. That is one caption with a thesaurus attached.
The six above are six different jobs. A question is asking for replies. A list tease is buying a rewatch. A plain descriptive line is written to be found in three months.
Pick the shape by what you want back, not by which one sounds best when you read them in a row.
One rule cuts across all six: do not spend the caption repeating your opening line. The viewer is already hearing it. A caption that echoes the first three seconds has used its only slot to say something twice.
The contrarian shape carries a bill. Take the other side of the usual advice and the replies will ask you to back it, so post it only when you can.
And the ask goes at the end, once. Four asks in one caption is no ask at all.
The six shapes and the job each one does
None of these is the best one. The right shape is the one that matches what you want the viewer to do.
| What the shape does | When it is the right one | |
|---|---|---|
| One-line hook | Extends the spoken opening instead of repeating it | When the first second of the video already does the work |
| Question | Asks something a stranger can answer in four words | When you want replies more than saves |
| Statement plus CTA | Names what the video gives, then asks once | When there is a link, a keyword or a booking behind it |
| List tease | Names the count and withholds the items | Videos that genuinely deliver a numbered list |
| Contrarian | Takes the other side of the advice everybody repeats | When you can defend it in the replies |
| Plain descriptive | Says what the video is, in words somebody would search | Evergreen how-tos and anything you want found later |
A caption cannot save a video, and there are two ways it loses one
Nothing typed in the post box rescues a video nobody watched. The first frame and the first spoken sentence do that job, and no caption competes with them.
What a caption can do is lose one, in two ordinary ways.
The first is the hashtag block. Hashtags sit inside the same 2,200 characters as your words, and hashtags placed first push the sentence behind the fold. Put them after the caption.
The second is asking for four things. Follow, comment, save, link in bio. Every extra ask lowers the odds of the one you actually wanted.
There is a quieter third: writing nothing at all. The caption is text, and text is the part of a post a search index can read, which makes the plain descriptive shape worth more than it looks.
The words a person would type into the search bar are the words worth having in the caption. Not stuffed. Once, in a sentence that reads like a person wrote it.
- One ask per caption. The extra ones cost the first one.
- Hashtags after the words, never in front of them.
- Say the caption out loud before posting. If you would not say it, do not type it.
- Put the searchable words in early, because the visible line is where they help.
From the caption to the video underneath it
The caption is the last thing you write. The video is the thing it sits under, and that is the part that takes an afternoon.
Cutroom turns one spoken take of up to three minutes, filmed on a phone, into a finished 9:16 video. It transcribes every word, trims the silence, lands b-roll on exactly the phrases you highlight, and captions the whole thing word by word from your own audio.
Those on-screen captions are the other meaning of the word from the top of this page. They come out of the transcript rather than out of a template, which is why they land on the beat your voice actually put them on.
The editing is directed on the transcript instead of a timeline. Delete a line you do not want and the cut rebuilds around it. Change the pace and the whole thing re-cuts.
The pricing is one sum. A batch is 100 credits and the export is 20 credits per output minute, so a 60-second finished video comes to about 120 credits.
Three finished videos are free inside your first seven days, no card, on the 600 credits granted at signup, and they export with a Cutroom mark across the middle. Lite is $19.99 a month for 1,250 credits, about ten finished one-minute videos and more when they run shorter, with cancelling one click away under Settings.
Questions people ask
- What does a TikTok caption generator actually write?
- The post text: the line that sits in the feed beside your username, under the video. It is not the subtitles burned over the footage, which are timed to your voice and made in an editor. If those are what you came for, the previewer at /tools/caption-style-preview plays your own sentence in five subtitle styles.
- How long can a TikTok caption be?
- The box accepts 2,200 characters, up from the 150 it used to allow. What the feed shows is roughly the first line, about a hundred characters, before it hides the rest behind a “more” link. Write for the hundred and treat the rest as a bonus for whoever taps.
- Do hashtags count toward the character limit?
- Yes. Hashtags live inside the same 2,200 characters as the words, which is why a block of them placed before the caption pushes your sentence behind the fold. Put them after the caption so the visible line stays yours.
- Should my caption repeat what I say in the video?
- No. The viewer is already hearing your opening line, so repeating it spends your only visible line saying the same thing twice. Extend it instead: ask the question the video raises, name the count in the list, or say the part you deliberately left out.
- Is what I type sent anywhere?
- No. The writer runs in your browser on this page, with no account needed. Nothing you type is uploaded or stored. Close the tab and it is gone, which is also why the captions come from hand-written patterns rather than from a model on a server.
- Will a good caption get the video more views?
- A caption cannot make a video that nobody watched perform. What it does is keep the people who did watch, ask them for one specific thing, and put searchable words on a post that otherwise carries none. Those are real jobs, and they are the ones worth writing for.
Write the one line that shows, put a single ask under it, and let the video do the rest.