Guide
Fourteen tested creatives per winner. Now count backwards
Work back from the win rate rather than guessing at a number. Published benchmarks put winners at roughly 5 to 8 percent of creatives. Expect about fourteen tested creatives per winner. This page covers the three numbers that size your month, what actually counts as a separate creative, and what one attempt has to cost. By the end you will have a monthly target you can defend in a meeting.
- Tested creatives per expected winner
- 14
- Published winner share across Meta ads
- 5-8%
- Numbers you need to size the month
- 3
Three numbers size the month, and you already have two of them
The win rate. Take it from published benchmarks until you have your own. Motion's analysis of 550,000+ Meta ads puts winners at roughly 5 to 8 percent of creatives.
The replacement interval. That is how long your current winner lasts before frequency climbs and cost per result follows. Read it off your own account rather than a benchmark.
The volume floor. That is how many impressions each creative needs before you can call the result. Below it a creative was not tested. It was merely published.
Put those together and the month sizes itself. At the middle of the published range you need about fourteen tested creatives per winner. A winner every six weeks means roughly ten tested creatives a month.
Run that with your own replacement interval before you accept anybody else's number, including this one. An account with a three month winner lifespan needs a fraction of what a weekly-refresh account needs.
Fourteen tested creatives, one expected winner
At the middle of the published 5 to 8 percent range. Every one of the fourteen costs full price, which is why cost per attempt decides the plan.
Published counts are worthless because a creative is not a unit of anything
Advice like run five creatives per ad set collapses on contact. A creative can mean a genuinely different concept. It can also mean the same ad with a different caption colour.
Two ads that differ in the opening three seconds are close to two different ads, for a viewer who never saw more than three seconds. Two ads that differ in the ending are nearly the same ad.
So count distinct concepts rather than files. A concept is a different reason to buy, a different objection answered, or a different person speaking.
Within a concept, count openings separately. The opening is what most of the audience ever experienced.
That distinction changes the plan. Ten files from three concepts is a smaller experiment than it looks. Three concepts with three openings each is a larger one.
What actually counts as a separate creative
Count concepts and openings. Everything below the line is a variant, and variants produce variant-sized differences.
A different concept
New reason to buy
A different opening
Most viewers saw only this
A different length or pace
Real, smaller effect
A different caption colour
Not a creative
This is the whole editor
Highlight a phrase and a clip lands on those exact words. No timeline, no keyframes, no layers.
The binding constraint is cost per attempt, not budget and not ideas
Most teams have more ideas than they can produce. What caps them is what each attempt costs in money, coordination and calendar.
Price the whole attempt rather than the invoice. Thinking hours, capture, assembly, coordination, versioning. Two of those five never appear on a bill, and they are usually the largest.
Then divide your monthly production spend by the number of creatives that got meaningful budget. That is cost per tested ad. Multiply it by fourteen for cost per winner.
Run that multiplication once before you set a target of ten a month. For teams commissioning individually, the number is often the reason the target was never met.
The route matters more than the effort here. Compare production routes on turnaround, revision rounds, cost per edit and consistency. Those four decide the count you can sustain.
Attempts a month, by what one attempt costs you
Your own arithmetic, not a benchmark. Halving the cost of an attempt does more for the count than doubling the ambition.
- A day of coordination each3 a month
- A briefing and two revisions6 a month
- One take, one batch, one export12 a month
Batch the recording, vary the opening, and stop scheduling shoots
The practical move is one recording session producing several ads rather than one session per ad. Record the body once. Then record six different openings against it in the same sitting.
Six shapes, not six rewordings. A contradiction, a named audience, a specific number, a demonstration, a confession, a question the viewer already asked.
That single habit changes the count more than any tool. The expensive part of a shoot is starting it rather than continuing it.
Cutroom removes the assembly line and leaves the rest to you. One take of up to three minutes returns a finished 9:16 MP4 you direct on the transcript rather than a timeline.
A batch is 100 credits. An export is 20 credits per output minute. A thirty second ad is 110 credits. Basic carries 2,500 credits a month, roughly twenty-two ads if you never re-export.
The trial is 7 days and 300 credits with no card. That is enough to find out whether your own count changes before any money is involved.
Set the target against the route you have, then make the route cheaper
Every route to volume has a cost no tool absorbs. Recording yourself costs willingness. Commissioning costs money and days. Customer content costs predictability.
So set the target against the route you actually have. A plan built on a route nobody will run dies in week two and takes the testing discipline with it.
Then work on what one attempt costs. That number moves the target more than ambition does. Assembly is the part of it with a published rate.
A batch is 100 credits. An export is 20 credits per output minute. Ten thirty second ads from five takes is 600 credits, against the 2,500 Basic carries at $39.99 a month.
Three facts to plan around. Three minutes of source per take. 9:16 output. A word-level transcript as the surface you edit on.
A concept needing a second location or an unfilmed demonstration goes to a videographer. Run the cheap route for the volume around it. That combination produces more tested creatives than either one alone.
Questions people ask
- Is there a minimum number of creatives to launch with?
- Enough to give each one meaningful volume, which for most small accounts is two or three rather than eight. Splitting a small budget across many arms produces many underpowered results, and underpowered results are worse than no test.
- Does more creative always help?
- Up to the point where each one still gets enough impressions to be read. Past that you are diluting rather than sampling, and the account learns less from twenty starved creatives than from four funded ones.
- How do I know my own win rate?
- Log every tested creative with one line: what it was, what it cost, whether it carried spend. After thirty entries the rate is yours rather than borrowed, and it will differ from any published range.
- Is volume the right move for every account?
- It pays once the offer has been tested. Volume multiplies whatever the offer already does, so a weak offer buys fourteen expensive confirmations. Fix the offer first, then buy attempts.
Count concepts, not files. Then price one attempt and multiply it by fourteen. In Cutroom that attempt is 110 credits, and it is the only figure on the list that does not move.