For your team
Your catalogue is not under-shot. It is under-sampled.
Every route to ecommerce creative bills per finished asset. So an invoice decides how many angles you test, not your ideas. That is why the account has run the same three hooks since March. Here is the arithmetic behind it, which angles the quote killed, how one phone take becomes a finished ad, and how many a month Cutroom puts in reach.

- Share of creatives that carry spend, per Motion
- 5-8%
- One finished 30 second vertical ad
- 110 credits
- What 2,500 credits on Basic comes to
- 22 ads
Six creatives a month against a 1-in-13 rate is a lottery with six tickets
Published benchmarks put the share of ad creatives that become winners at roughly 5 to 8 percent.
The figure comes from Motion's analysis of 550,000+ Meta ads. Call it one in thirteen to one in twenty.
Now hold six against that. Six is not a creative problem.
It is a sampling problem, and no brief written on a Sunday night fixes one.
The creative review never has that number on the agenda. It has the six ads on the agenda.
The argument is always about which hook was weak.
Blaming the hook is comfortable, because the hook is fixable by Thursday.
Sample size is fixable only by changing what one attempt costs.
That is the number this page is about. Move it and the rest of the account changes shape.
Twenty-six creatives at the published rate
Two carry the spend. Shipping six a month means drawing six tiles from this grid and calling it a testing programme.
Nobody rejected the returns-policy angle. The quote did.
Open the last brief and count the angles that died before anything was filmed.
The founder-story open. The one built on your returns policy. The one aimed at a buyer replacing something that broke.
None of those were rejected on evidence. They were rejected on price.
Production quotes per asset. At a few hundred dollars each you commission only the ones you were already confident about.
So the account has been testing your taste. It has not been testing your market.
The angles you dismissed never got to be wrong in public, and public is the only place the data lives.
There is a second cost nobody books. An ad that cost six hundred dollars has a defender in the room.
A dead creative runs until Friday. An attempt that costs under two dollars dies on Wednesday with no meeting.
At catalogue level it compounds. A brand with sixty SKUs usually has video for eight.
Those eight are the products that were already selling, which is how a bestseller list becomes self-fulfilling.
- The long tail of the catalogue carries no creative, because nobody books a shoot for a twenty-four dollar accessory
- Killing losers fast is most of what a testing programme is, and sunk cost is what stops it
- The unit that matters is angles per SKU, not videos per SKU
This is the whole editor
Highlight a phrase and a clip lands on those exact words. No timeline, no keyframes, no layers.
The transcript is the control surface, so a variant is a decision not an afternoon
Upload one talking-head take, up to three minutes.
Cutroom transcribes the read, drafts the cut, places b-roll against the words and returns a finished vertical MP4 with captions burned in.
Not a project to assemble. A file you upload to the ad account.
You direct it on the transcript. Highlight a phrase and a clip lands over exactly those words.
Delete a line and the cut rebuilds around what is left. Change the pace and it re-cuts the whole thing.
The mechanism matters because of what it removes. No timeline to scrub, no keyframes, no layers.
Version four is a judgement about the ad rather than an evening in software.
The compiler owns every timing decision. Cut a sentence and the captions, the clip placement and the cut points all come back consistent.
None of them were placed by hand to begin with, so there is nothing to chase down the track.
- Swap the clip in any spot, from your own library or from millions of free ones
- Emphasise a word so it pops in the captions the way it popped in your mouth
- Change how each clip enters, using Cut, Whip, Punch, Glitch or Sweep
- Six caption packs, plus colour, weight, size and position
One phone take to a file in the ad account
One take
Up to 3 minutes, filmed on a phone
Batch
100 CR: transcript, director pass, b-roll
Direct on the words
Highlight, delete a line, set the pace
Export
20 CR per output minute, 9:16 MP4
At 110 credits an attempt, the shoot day yields eleven reads instead of one
A batch is 100 credits. That covers transcription, the director pass and the b-roll search.
Rendering costs 20 credits per output minute, so a thirty second ad adds 10. Call it 110 credits a finished ad.
Basic is $39.99 a month for 2,500 credits, about 22 finished ads.
Premium is $79.99 for 5,000 credits, about 45. A $15 top-up adds 1,000 credits when a launch week runs hot.
Watch what that does to an afternoon with a camera.
The same hour used to produce one deliverable and a fortnight of waiting.
Now it produces eleven reads, and the eleventh costs what the second cost.
The account stops running on the ideas your team defended in a meeting.
It starts running on the ideas buyers responded to. Those are usually two different lists.
Finished thirty second ads a month, by plan
At roughly 110 credits an ad. The trial is 7 days and 300 credits with no card.
- Trial, 300 credits2 ads
- Basic, $39.9922 ads
- Premium, $79.9945 ads
- Each $15 top-up9 more ads
Your product footage, your library, and the one file that comes out
The b-roll search covers millions of free clips. Kitchens, hands, mornings, the world around your product.
Your own footage sits beside it. Film the item in a customer's hands once, twenty minutes on a phone.
Upload that clip to your workspace library and it is available to every ad you make afterwards.
Highlight the phrase where you name the product and the shot lands over exactly those words.
Swap the clip in any spot, from your library or from the free ones, without leaving the editor.
The flat facts to plan around. Exports are 9:16 MP4, which is the shape Reels, TikTok and Shorts all want.
Source takes cap at three minutes. Each project produces one finished ad.
Somebody has to speak on camera. A founder, a warehouse manager, a customer you paid for twenty minutes.
That requirement is also the reason the output performs. A real person making a trust-led claim is the ad that carries spend on an ecommerce account.
Questions people ask
- How many finished ads does Basic actually cover?
- About 22 thirty-second ads. Each is 100 credits for the batch plus 20 per output minute to render, so roughly 110, against 2,500 credits for $39.99 a month.
- Can my own product footage appear in the ad?
- Yes. Upload your clips to the workspace library, then highlight the words where you name the product and the shot lands over them. The free b-roll search covers everything around it.
- Do I need a studio or an on-camera person?
- A phone and a quiet room. If nobody in the company will film, the avatar module turns a photo and a voice-over into a lip-synced presenter at 320 credits per minute of render.
- Who should not buy this?
- Brands whose ads are built on product cinematography and a colour grade with nobody speaking. Book an editor for those. Everything vertical and spoken is faster and cheaper here, at about 110 credits an ad.
One phone take in. A finished 9:16 ad out, b-roll on the right words, captions burned in, for about 110 credits. Most tools hand you a project file and call that the product.