One take in, one finished ad out, about 110 credits

It needs one talking-head video of up to three minutes. Filmed on a phone is fine and usually better. No footage library, no editing experience, no card for the trial.

It makes a finished 9:16 MP4 with captions rendered into the picture, footage cut to the words it covers, an opening headline and music underneath if you want it.

One trip through the machine is 100 credits. That covers transcription, the directing pass and the footage search. Export is 20 credits per minute of finished video.

So a thirty second ad, made and downloaded, is about 110 credits. The 7 day trial gives you 300 credits and asks for no card, which covers two of those end to end.

Basic is $39.99 a month for 2,500 credits, around twenty finished ads. Premium is $79.99 for 5,000. A $15 top-up adds 1,000 credits whenever a month runs long.

Nothing else is required to start. No brand assets to prepare. No template to choose. No stock library subscription underneath it.

The first cut arrives finished, which is where most tools stop. Then you change it by marking your own words.

Highlight a phrase and a clip lands over exactly those words. Swap that clip, or search millions of free ones. Delete the line that went nowhere and the cut rebuilds around the gap.

Change the pace and the read re-cuts. Emphasise the word carrying your offer. Trim the opening. Change how clips enter. Restyle the captions. Rewrite the headline. Set the music level.

Every instruction points at a word rather than at a second, so an edit above your mark cannot move it.

What is missing from that list is the afternoon. No timeline. No scrubbing. No clip handles. No caption timing done by hand at eleven at night.

Motion looked at more than 550,000 Meta ads. Roughly 5 to 8 percent of creatives become winners. Call it one in thirteen to one in twenty.

Take that seriously for a minute. Six ads a month is not a testing programme against those odds. It is a coin flip with a slow denominator.

The instinctive response is to try harder on each attempt. Better brief, better creator, another revision round. That raises quality a little and cuts attempts a lot.

The arithmetic gets worse, and it gets worse quietly, over a quarter, while everybody feels productive.

The response that moves the number is to make each attempt cost less. Not worse. Fewer hours between deciding to test something and looking at the result.

Making one ad is not the hard problem. Every tool on the market can make one ad.

The hard problem is making the ninth one differ from the eighth by a single sentence, in eight minutes rather than ninety, without breaking anything downstream.

That works here because your edit is a short list of instructions applied to the original take. It is not a pile of accumulated changes to a video file.

Delete a line and the whole cut is computed again. The clip after it does not slide. The captions do not drift. The bed does not stop early.

So the ninth export costs 10 credits and the batch is not charged twice. That is the only reason a team of two can put thirty creatives into an account in a month instead of six.

Watch what teams do once the ninth version is cheap. They stop holding a meeting about whether the second hook is worth testing, because testing it takes eight minutes and 10 credits.

A hired editor wins two rows. They tell you the hook is weak before anybody films it, and they work from sources far longer than three minutes.

Judgement before the shoot is worth paying for. The machine cuts what you said and has no opinion about whether it was worth saying.

So use both. Buy the judgement once, on the script. Then make the twenty versions here for 10 credits each.

Nothing predicts which ad wins. Nothing can, and that is the entire reason volume beats polish.

The scope here is one take of up to three minutes, out as a 9:16 MP4, with removal working by the line. Those follow from one decision.

This is a directing tool rather than an editing suite. Directing decisions go in, editing decisions stay out, and an eleven item move list is why version twenty needs no repair.