Guide
How many ad creatives to test: about 14 a month for one winner
Updated
Test about 14 ad creatives a month to get about one winning ad. With only 6 a month, there is a 65% chance the month ends with no winner.
In short
- If you test about 14 ad creatives a month, you can expect about one winning ad.
- Only 5 to 8% of ads become winners, according to Motion's analysis of 550,000+ Meta ads.
- With 6 ads a month, there is a 65% chance the month ends without a single winner.
- With 14 ads that chance falls to 36%, and with 30 ads to 11%.
- The cheapest way to test more is to film one video and make several versions of it.
- Share of Meta ads that become winners, per Motion
- 5-8%
- Ads a month for about one winner, at a 7% win rate
- 14
- Chance that 6 ads a month give no winner at all
- 65%
Why test about 14 ad creatives a month?
Only a small share of ads become winners, so you need enough of them to get one.
- Motion, an ad analytics company, studied 550,000+ Meta ads and found that 5 to 8% became winners.
- Motion counts an ad as a winner when it spends at least 10 times as much as a typical ad in the same account, and at least $500.
- A 5 to 8% win rate means about 1 winning ad in every 13 to 20.
- At 7%, near the middle of that range, 14 ads give you about one winner (14 × 0.07 = 0.98).
- Your own win rate will be different. Plan with 5 to 8% until you have tested about 40 ads of your own.
What a 7% win rate looks like
Fourteen ads, one winner. At a 7% win rate this is the normal result, not a bad month.
What are the chances of a month with no winning ad?
The fewer ads you test, the more likely a whole month passes without a winner. At a 7% win rate, 6 ads a month leave you with no winner 65% of the time.
- With 6 ads a month, you can expect less than half a winner.
- With 14 ads a month, you can expect about one winner, and the month ends empty 36% of the time.
- With 30 ads a month, you can expect about two winners, and the month ends empty only 11% of the time.
Ads per month and the chance of no winner
At a 7% win rate, each ad has a 93% chance of not winning. The chance of no winner is 0.93 multiplied by itself once for every ad.
| Ads tested a month | Winners you can expect | Chance of no winner |
|---|---|---|
| 6 | 0.42 | 65% |
| 14 | 0.98 | 36% |
| 20 | 1.4 | 23% |
| 30 | 2.1 | 11% |
Ads tested a month: 6
- Winners you can expect
- 0.42
- Chance of no winner
- 65%
Ads tested a month: 14
- Winners you can expect
- 0.98
- Chance of no winner
- 36%
Ads tested a month: 20
- Winners you can expect
- 1.4
- Chance of no winner
- 23%
Ads tested a month: 30
- Winners you can expect
- 2.1
- Chance of no winner
- 11%
This is the whole editor
Highlight a phrase and a clip lands on those exact words. No timeline, no keyframes, no layers.
Is a month with no winning ad a bad sign?
No, not on its own. With 6 ads a month, a month without a winner is the most likely result.
- Two empty months in a row happen 42% of the time when you test 6 ads a month.
- A whole quarter without a winner happens 27% of the time.
- So one empty month is not a good reason to fire your agency or change your brand look.
- If you change them and the next ad wins, you cannot tell whether the change helped or you were just lucky.
- Judge your testing after a set number of ads, such as 14 or 20, not after a set number of weeks.
Empty months at 6 ads a month
At a 7% win rate: 0.65 multiplied by itself once per month.
| No winner for | Chance |
|---|---|
| 1 month | 65% |
| 2 months in a row | 42% |
| A whole quarter | 27% |
No winner for: 1 month
- Chance
- 65%
No winner for: 2 months in a row
- Chance
- 42%
No winner for: A whole quarter
- Chance
- 27%
How much budget do 14 ads a month need?
Each test ad needs a daily budget of at least your target cost per sale. Less than that, and it cannot bring in even one sale a day.
- With 3 test ads running at once, you need about 3 times your target cost per sale every day.
- At a $30 target cost per sale, that is $90 a day, or about $2,700 a month.
- Launching 3 new ads every week and switching off the losers comes to about 13 ads a month, close to 14.
- Keep the promising ads running until they have enough sales to judge. Only the clear losers go off after a week.
- On a smaller budget, test fewer ads. Judge them first on cheaper numbers, such as clicks or how many people watch the first 3 seconds.
Daily test budget with 3 ads running
A rough check over 30 days. Meta will not split the money evenly between the ads.
| Your target cost per sale | Budget per day | Budget per month |
|---|---|---|
| $20 | $60 | About $1,800 |
| $30 | $90 | About $2,700 |
| $50 | $150 | About $4,500 |
Your target cost per sale: $20
- Budget per day
- $60
- Budget per month
- About $1,800
Your target cost per sale: $30
- Budget per day
- $90
- Budget per month
- About $2,700
Your target cost per sale: $50
- Budget per day
- $150
- Budget per month
- About $4,500
How can you test more ads without filming more?
Film one video and turn it into several versions. Each version changes one thing, so you can tell which change made the difference.
More ads from one video
Film one video
You talking to the camera about one offer
Make versions
Change only the opening line, the offer or the price you lead with
Launch two or three
Give each one a daily budget of at least your target cost per sale
Switch off the losers
Then launch the next two or three versions
Where Cutroom fits
Cutroom edits the video for you, so another version of an ad is a quick text edit and a new export.
- You upload one clip of yourself talking, up to 3 minutes long.
- Cutroom cuts it, adds captions and b-roll, and gives you a finished 9:16 MP4 in about a minute.
- You make a version in the editor: write a new headline on screen, pick another caption or hook theme, or move the b-roll to other words.
- A 30-second ad costs 110 credits. Another version of the same video costs only its export, 10 credits.
- Your first video is free, with no card, and you have 7 days to make it.
What it costs in Cutroom
| Item | Cost |
|---|---|
| Edit (cuts, captions, b-roll) | 100 credits |
| Export | 20 credits per minute of video |
| A 30-second ad | 110 credits |
| Another version of the same 30-second ad | 10 credits |
| First video | Free, no card, within 7 days |
| Lite plan | $19.99 a month for 1,250 credits |
Item: Edit (cuts, captions, b-roll)
- Cost
- 100 credits
Item: Export
- Cost
- 20 credits per minute of video
Item: A 30-second ad
- Cost
- 110 credits
Item: Another version of the same 30-second ad
- Cost
- 10 credits
Item: First video
- Cost
- Free, no card, within 7 days
Item: Lite plan
- Cost
- $19.99 a month for 1,250 credits
Questions people ask
- How many ad creatives should I test per month?
- About 14, if you want about one winning ad a month. That assumes a 7% win rate, near the middle of the 5 to 8% that Motion found in 550,000+ Meta ads.
- Is a 5 to 8% win rate something I can rely on?
- For planning, yes. It is a benchmark from Motion's data, not a promise. After about 40 tested ads, plan with your own win rate instead.
- What if I cannot make 14 ads a month?
- Make several versions of one video instead. Film it once and test a different opening line on each version. Each extra version costs far less than a new shoot.
- How much budget do I need to test 14 ads a month?
- About 3 times your target cost per sale a day, if you run 3 test ads at once and launch 3 new ones each week. At a $30 target that is $90 a day, or about $2,700 a month.
- Does testing many ads hurt Meta's learning?
- It can, if you run them all at once on a small budget. Launch two or three at a time, give each one at least your target cost per sale a day, switch off the losers, then launch the next few.
- How long should an ad run before I judge it?
- Until it has enough sales to judge, and decide that number before you launch. At five sales a day, two days gives you ten sales, which is too few to tell a winner from luck.
- How do I work out my own win rate?
- Divide your winning ads by all the ads you launched. Keep a list of every ad you test and whether it became a winner. After about 40 ads, the number is reliable enough to plan with.
Film one video and see how many ads it can become. Your first video is free.