From your long video to finished Shorts in five minutes.
Drop your long video. Tick the moments in the transcript. Get the whole batch back, finished.
Framing, captions, hook: all automated.
5 free minutes of finished clips every month. No card. No watermark.
A real clip, straight off the line: hook up top, captions word by word. The person is a generated avatar.
How it works
One decision for a whole batch
Drop your long video: a podcast, a stream, an interview. Go do something else. By the time you're back, the transcript is waiting.
You read it, and you take the moments worth taking. That's all you do: pick, in text. No timeline to scrub frame by frame, no cursor to drop on the exact second.
You set the look once for the whole batch, you hit go, and you get vertical videos back: edited, framed, captioned, ready to post.
You make the calls. You don't edit a thing.
You drop your long video
Up to two hours of source. The quote tells you what it costs before anything starts.
You pick in the transcript
One tap takes a sentence, a tap on the next one extends the stretch.
You set the look, once
One framing and one caption style for the whole batch, not for a clip.
You collect your clips
Vertical MP4s, ready to post, no watermark.
5 free minutes of finished clips every month. No card. No watermark.
The problem
AI automated the wrong half of the job
Two years ago, getting a clip out clean took skill: reframe it vertical, caption it, sync it, export it. That's exactly the half AI took. Today anyone gets a captioned clip out in three minutes.
What used to be rare is now free, so everyone does it: this morning, ten of you started from the same video, ran it through the same tool, and got the same clip. Looking like everyone else used to cost nothing. Now it does.
So making the clip doesn't set you apart anymore. The moment you pick does.
Now look at what your tool automates. It picks the moments for you, the one thing that's yours. And it leaves you the finishing to redo clip by clip, the one thing that should be automatic.
The automation is on the wrong side.
The landscape
Three ways to make clips today
An editor makes you do everything. An AI clipper picks for you.
AI clippers automate discovery: they go hunting for your moments. ClipFinish automates production: you pick the moments, the machine runs the batch. Two tools, two ways of working.
| The approach | Who picks | Who does the work | What's left for you |
|---|---|---|---|
| The editorCapCut, Premiere | You | You | The same move, clip after clip |
| The AI clipper"find my viral moments" | The AI | The AI | A batch to redo clip by clip |
| The production lineClipFinish | You | The machine | Nothing |
The call
You pick the moments. The batch follows.
Either the AI picks for you: it starts mid-thought, cuts before the punchline, catches the laugh instead of the line that caused it. Or you spend half a day in a timeline hunting for moments you already had in your head. Neither works, because the real problem is somewhere else: finding a good clip is a gut call, not a technical job.
In ClipFinish you pull your clips straight out of the transcript. You click passages of text, you play them back, you preview them, you keep them.
And the layout belongs to the batch, not to the clip. AI clipping apps change the framing from one clip to the next because their tool treats every clip as a separate video: it has no idea a batch exists, so it re-decides every time. Here, no clip can fall out of the batch. It's a rule baked into the app, not a box someone could forget to tick.
Same with the hook: you write it yourself, or you let our AI draft one if you can't be bothered. Either way, it's your call.
Nothing that starts mid-sentence. Nothing you have to open to fix.
On the public feedback board of one of the biggest tools out there, the top-voted request, 155 votes, open since 2023, is exactly this: pick your own moments instead of taking whatever the AI hands you. On the same board, 56 paying subscribers want the framing to stop changing from clip to clip, and 62 more just want to move the captions on all their clips at once.
Captions
Captions you can read where people actually watch
Other tools drop the text "at the bottom" and that's that. So on a phone, the caption, the icons and the account name land right on top of it. You find out after you posted.
In ClipFinish, the zones where TikTok and Instagram put their interface are reserved up front: your captions can't go there, not on one clip, not on any of them. And they follow words, not sentences, so each word shows up as it's said.
Nothing to move up, nothing to re-sync.
On a campaign, captions aren't a matter of taste. They're a condition of getting paid.
- Platform interface
- Buttons and counters
- Caption and account name
- Where text is allowed
- Hook · Captions
Credit
What your credit actually buys
The real question about credit isn't the price. It's what it buys.
Elsewhere it goes into things that give you nothing back: suggested clips you throw away, renders that crash, and a refund you have to chase by email with a project ID, a timestamp and an error message.
In ClipFinish, analysing your video costs one minute for every five you drop. That's an advance, not a surcharge: up to that much, the clips you cut from it aren't charged. Past it, they count for how long they run.
And if a clip fails for good, its minutes come back to you. On their own, right away, without you writing to anyone. One failed clip doesn't take the rest down: the others come out anyway.
You never pay twice for the same second.
Straight up
What we have, what we don't, what we don't do
What we have
Every sentence on this page describes what the product does today, not what's planned.
What we don't have yet
Customer testimonials. ClipFinish just launched and has no customers to quote. We'd rather tell you than make up three reviews with first names and headshots.
So why trust us?
Trying costs you nothing. You get 5 minutes of finished clips every month, no card, no watermark, and you judge for yourself. That's a day of production. Enough to know, and you owe us nothing.
5 free minutes of finished clips every month. No card. No watermark.
What ClipFinish doesn't do, and why
- There's no timeline
- That's not something missing: a timeline forces you to think clip by clip. We built ClipFinish for working in batches.
- Two caption styles, not twenty
- Two styles tuned to the pixel beat twenty that let bad clips out.
- We don't post for you, and we don't schedule
- That's not an oversight: auto-posting wants business accounts, and yours may not be, it disconnects every couple of months, and no outside tool can post with a trending sound. We'd rather hand you a finished batch and let you post it the way you know how.
- We don't pick your moments
- That's where you're good. No AI of ours is ever going to beat you at it.
Before you start
FAQ
- What languages can I clip in?
- Transcription works out the language on its own. Rendering only draws the Latin alphabet. A source in another script is refused clearly, not rendered as boxes.
- What resolution do my clips come out in?
- 1080 x 1920, what TikTok, Reels and Shorts expect. The frame rate is taken from your source, up to 60 frames a second.
- What files can I upload?
- The usual video files, MP4 and MOV first among them. Your source can run up to two hours, and each clip up to three minutes.
- How long do you keep my files?
- 3 days on the free plan, 30 days on the paid ones. Download your clips to keep them longer.
- Is there a watermark on my clips?
- No. Not on any plan, free included.
Pricing
What it costs
Free
free
5 minutes of finished clips a month
About 10 thirty-second clips
Enough to start, no rush.
Files kept 3 days
Solo
€19a month
75 minutes of finished clips a month
About 150 thirty-second clips
Enough to post every week.
Files kept 30 days
Pro
€39a month
250 minutes of finished clips a month
About 500 thirty-second clips
Enough to post every day.
Files kept 30 days
The moment you pick is the only thing nobody can copy.
The rest, ClipFinish does for you, so you get your time back.
5 free minutes of finished clips every month. No card. No watermark.