Comparison
Permutations run out. A person saying something does not.
Static generators permute headline, image and layout across every size a platform sells. AdCreative does that at real volume. Seventy-two arrangements of one claim is still one claim. Here is where permutation stops paying, what a new angle costs, and how Cutroom produces one in four minutes.
- asset type Cutroom produces
- 1
- per finished minute exported
- 20 CR
- of creatives carry the spend
- 5-8%
Twelve placements, four sizes each, and a deadline: static generation wins that
Display, social and retargeting all want different dimensions of the same idea. Producing them by hand is dull, slow work.
A generator that permutes headline, image, colour and layout across every required size removes a genuine bottleneck.
Scoring adds something too. When forty variants exist, ranking them is more useful than staring at a contact sheet.
There is a workflow argument too. Statics get approved in a document, so the review chain never waits for anybody to watch anything.
For placement coverage across a platform, that is the right buy and nothing here replaces it.
Here is the ceiling. Six headlines by four images by three colourways is seventy-two files and one concept.
The feed prices concepts, not files. Once the audience has met the idea, variant seventy-two is fatigue in a new colour.
Cutroom makes new concepts instead. Film forty seconds answering a different objection and upload it.
Back comes a finished 9:16 MP4, captions burned in and coverage sitting on the words you marked.
Permutation multiplies variants of one idea. It does not produce a second idea.
Permutation stops paying faster than expected. The variation sits in the layer a viewer stops noticing first.
Vary a layer, not a setting. A different font is not a different ad.
Video with a person in it has a different failure curve. A new angle, a new objection answered, a new opening claim is a new idea rather than a new arrangement.
A different image over the same claim is barely one either.
Statics and talking-head video answer different halves of the same volume problem.
The practical symptom is a plateau. Cost per result improves for two weeks, flattens, and then no amount of new colourways moves it.
At that point the account needs a different thing to say, not a different way to arrange the same thing.
Thirteen creatives, one carrying the spend
Motion's analysis of 550,000+ Meta ads puts winners at 5 to 8 percent. Seventy-two permutations of one idea is still one idea in that grid.
This is the whole editor
Highlight a phrase and a clip lands on those exact words. No timeline, no keyframes, no layers.
Mark the transcript, and a new angle costs four minutes rather than a shoot
Film forty seconds answering one objection. Upload it, up to three minutes. It comes back transcribed, drafted and captioned as a 9:16 MP4.
Highlight a phrase and a clip lands over exactly those words. Swap it, or search millions of free ones.
Delete the line that missed and the cut rebuilds. Change the pace and the whole piece re-cuts.
Emphasise the word the claim turns on so it pops in the captions. Fix a mis-heard word and the text follows.
Write a different opening headline and export again. Same take, different first three seconds, 20 credits a minute.
That is the closest thing here to permutation, and it operates on the layer that changes performance.
Coverage is capped by pace: 40 percent on chill, 45 on normal, 52 on fast. The claim keeps a person attached to it.
- Six caption packs, adjustable for colour, weight, size and position
- Coverage capped at 40, 45 or 52 percent by pace
- Batch 100 credits, and a second cut of the same take does not re-batch
- Somebody has to film, or own a photo the avatar module can drive
One asset type, one placement: the specification, written flatly
One talking-head take of up to three minutes goes in. One 9:16 MP4 comes out, captions burned in.
There is no static output and no image generation. Display sizes and square placements are outside this entirely.
There is no scoring model predicting which creative will win. Published scores describe somebody else's placement mix.
A person has to appear, or the avatar module needs a photo you own plus a voice-over at 320 and 70 credits a minute.
There is no timeline underneath and nothing of yours is stamped on the file.
There is no ad account integration. You download the file and upload it where it runs.
Those are the facts to plan around. The one number they do not touch is new ideas per month, which is next.
Cold traffic is where a new idea pays, and new ideas cost four minutes here
Winners run at 5 to 8 percent of creatives, per Motion's analysis of 550,000+ Meta ads. Volume beats perfection on both routes.
Talking-head video carries cold traffic, where a stranger decides whether a claim is worth three more seconds.
The volume lever there is the cost of a second attempt. Here it is 20 credits a minute and about four minutes.
So a new objection gets answered on camera the same afternoon somebody raises it.
One finished minute from a fresh take is about 120 credits. Seven days and 300 credits with no card covers two.
Keep the static generator carrying retargeting, display and every placement where video is the wrong load.
Film two claims for the cold side and read the cost per result yourself.
AdCreative and Cutroom, row by row
The top two rows go to AdCreative, and placement coverage should weigh them heavily. The rest is what permutation cannot reach.
| AdCreative.ai | Cutroom | |
|---|---|---|
| Statics at every placement size | Its whole purpose | 9:16 video only |
| Nobody has to film | Assets and copy in | A take, or your own photo |
| Cut rebuilds when a line goes | Static by definition | Delete, and it recompiles |
| Captions burned in for a muted feed | Nothing to caption | Six packs, one tap |
| A person making the claim | Layout and imagery | Your face, your read |
| New angle, not new arrangement | Same idea, restyled | New take, new objection |
| Footage tied to a phrase | Static by definition | Highlight, clip lands |
Questions people ask
- Does Cutroom generate any static ad assets?
- None. There is no image output, no banner sizes and no headline permutation engine. The only thing it produces is a 9:16 MP4 cut from a talking-head take under three minutes.
- Can I make many variants of one video ad?
- Yes, from one batch. Change the headline, the pace, the caption pack or which phrases carry coverage, then export again at 20 credits a minute. The take is transcribed once and the batch is not charged twice.
- Does Cutroom predict which creative will perform?
- No, and treat any tool that claims to with suspicion. A published score is built on somebody else's placements and audiences. The only reliable signal is your own account, which is an argument for more attempts.
- Who should not buy Cutroom?
- Anyone whose gap is static coverage across display placements. Buy the generator for that. Come here for the cold-traffic ads, where the claim needs a person attached to it.
Seventy-two arrangements of one claim is still one claim. Cutroom turns a second claim into a finished 9:16 ad in four minutes, coverage sitting on it.