Guide
A win rate is a fraction, and the bottom half is negotiable
Published benchmarks put winners at 5 to 8 percent of creatives. That is one in thirteen to one in twenty. Your own number is decided by the denominator. Most teams pick a flattering one without noticing. This page covers the three denominators, a definition of a winner that survives a quarter, and the two levers that move the fraction. By the end you will know which of the two is available to you this month.
- Published winner share, per Motion's analysis of 550,000+ Meta ads
- 5-8%
- The optimistic end of that range
- 1 in 13
- The pessimistic end of that range
- 1 in 20
The published range describes creatives that reached a real test
The benchmark to plan against is 5 to 8 percent. It comes from Motion's analysis of 550,000+ Meta ads. Anyone quoting a tidier figure than a range is inventing precision.
Read the clause carefully. It counts creatives that got enough spend to be judged. It does not count everything a team produced. Your raw production almost never matches that population.
So your measured rate can sit below the published one and nothing is wrong. You are counting a wider set of attempts than the benchmark did.
A win rate is comparable to itself and to nothing else. Track your own line across quarters. Use the published range only until you have a line to track.
Three winners, three denominators, and only one of them is checkable
Say forty creatives were made last quarter. Twelve got meaningful spend and three of those held your threshold.
Three out of twelve is 25 percent. That is the number that ends up on the slide. Three out of forty is 7.5 percent. That is the one you can act on.
The twenty-eight that never got funded were still work you paid for. They consumed briefs, shoots, edits and calendar. Excluding them tells you how good your funded creatives are. That is not a lever.
Keep both if you like. Label them. Wins over everything produced is production efficiency. Wins over everything funded is selection quality.
There is a third denominator nobody keeps. It is the most honest one. Count everything briefed. The concept that died in a document. The shoot that got cancelled.
Three winners, three denominators
Same quarter, same three wins. The bottom number decides which story you tell the board.
- Of 12 funded25%
- Of 40 produced7.5%
- Of 60 briefed5%
This is the whole editor
Highlight a phrase and a clip lands on those exact words. No timeline, no keyframes, no layers.
A definition written after the batch always flatters the batch
Ask five people what a winner is and you get five answers. The best click rate. The one the founder likes. The one still running.
A usable definition has four clauses. A metric that connects to money. A threshold. A volume floor. A duration. Miss one and the definition gets argued about on a Friday.
Here is a workable version. A creative wins if it holds target cost per acquisition across at least fifty conversions, for a set number of days rather than one spike.
Write it down before the batch goes live. A definition chosen afterwards flatters the batch. You will never notice you did it.
Click rate on its own makes a poor threshold. It is cheap to move and easy to inflate with a misleading opening. It belongs in elimination, not in the definition of a win.
The four clauses of a definition that survives a quarter
All written before launch, in one sentence. Every clause you leave out is a clause somebody argues about later.
A metric
One that connects to money, not to clicks
A threshold
A number, so the claim can be false
A volume floor
Fifty conversions, so a fluke cannot count
A duration clause
It has to hold, not only spike
Threshold drift and regression push your measured rate down over time
As spend grows, the cost you can accept falls. Yesterday's winner fails today's bar. Your rate drops without a single creative getting worse.
Regression does the rest. A creative judged on its first hundred impressions looks brilliant or terrible for reasons unrelated to the creative. About half of those verdicts reverse.
A worked case. Forty produced, three winners, 7.5 percent. Strip out the eight that never got past a first read and the rate rises to 9 percent on a smaller base.
Both numbers are true. One of them is useful for planning. Pick the denominator once, write it down, and stop changing it between meetings.
Then use the rate as a planning input rather than a school report. Multiply it by monthly output and you get expected winners a month. That is the only job the number has.
Forty produced, three that held the threshold
The other thirty-seven are not waste. They are the search, and the search has a published price.
Two levers, and only one of them is available this month
Raise the rate. That is slow. It comes mostly from better hooks and better offers. Or raise the volume. That is mechanical, and it is available today.
Cutroom is built for the second one. One take in. A finished 9:16 MP4 out. Directed on the transcript rather than assembled on a timeline. It changes what an attempt costs, and attempts are the half of the fraction you control.
Nobody else closes that gap. A timeline hands you control and a queue. A clip generator hands you footage you still have to assemble. This hands back the finished vertical file. Captions burned in. Footage on the words that needed it.
The facts to plan around. Somebody records a take of up to three minutes on camera. Every export is a 9:16 MP4. A batch is 100 credits. An export is 20 credits per output minute. The trial is 7 days on 300 credits with no card.
Then keep measuring the top half honestly. A rate stuck near 2 percent across forty creatives is telling you about the offer. No production speed answers that one.
Questions people ask
- Should a creative that wins for a week count as a win?
- Only if your definition says so. Add a duration clause: it must hold the threshold for a set number of days or a set spend. A creative that beats the bar for two days and collapses is usually an audience effect being credited to the creative.
- Do I measure win rate per creative or per concept?
- Both, and they answer different questions. Per creative tells you how efficient your production is. Per concept tells you how good your strategy is. A concept with five failed executions is often a good idea shot badly.
- How many creatives before my own win rate means anything?
- At a rate near 7 percent you need dozens before the estimate settles. Under twenty creatives your measured rate is mostly noise. Use the published range until your sample is well into the forties.
- Is a high win rate a good sign?
- Not always. It usually means the bar is too low or the variations are too timid. If nearly everything you ship wins, you are not exploring far enough to find the outlier that pays for the programme.
- Who should walk away once they have done this sum?
- Anyone whose measured problem is the offer rather than the count. Spend the quarter on pricing and proof first. If the count is the problem and nobody will film, book a creator and run their take through the same editor for 100 credits.
One in thirteen to one in twenty. The fixable half of that fraction is the number of attempts. One take in, one finished vertical ad out, and nothing else takes you the whole way.