The short answer
At $100 a day, run ten to fifteen creatives and never more than fifteen. Too many creatives on too little budget means none of them gets enough spend for Meta to learn from: the campaign stays in the learning phase and every ad is starved. One creative is the opposite mistake, because it gives the algorithm nothing to choose between. New creatives replace the ones Meta has stopped spending on rather than piling on top, so the live set stays at ten to fifteen as the account turns over.
The usual advice is to give Meta plenty to choose from, and at scale it is right. At the floor it is how a $100-a-day campaign spends a month in the learning phase. Fifty creatives on $100 a day is $2 a day each before Meta has picked anything, and $2 a day tells the algorithm very little about any of them.
The arithmetic at $100 a day
| Creatives live | Spend each, if Meta split it evenly | Qualified leads each a week, at $24.07 |
|---|---|---|
| 1 | $100 a day | About 29 |
| 10 | $10 a day | About 3 |
| 15 | About $6.70 a day | About 2 |
| 30 | About $3.30 a day | About 1 |
| 50 | $2 a day | Under 1 |
Meta does not split it evenly, and that is the point: within a day or two it puts about 90% of the spend on one or two ads. But it chooses on early results, and at $100 a day there are not many to go round — about 29 qualified leads a week at the $24.07 our main campaign measured. Spread across fifty ads, most of them never produce one before the choice is made. That is what the board means by nothing learns: the campaign sits in the learning phase and every ad in it is starved.
Too few is a mistake as well
The opposite error is one creative on $100 a day. Alex’s comparison is a fire: it needs oxygen to grow, and the oxygen is money and creatives together. One ad gives Meta nothing to choose between and nothing to move to when it fatigues, and every ad fatigues — even a winner is gone within a month or two. Ten to fifteen that complement each other is the balance at the floor; the six angles in a static and a video version each is twelve.
Running the fifteen
- Start with ten to twelve: the six angles in static and video, plus a few variants of anything that has already worked.
- Leave them alone for the first days while Meta concentrates the spend. The ads it is ignoring are not failures yet.
- When new creatives are ready, retire the ones Meta has stopped spending on to make room, rather than adding to the pile.
- Keep one campaign, one ad set and one event at this budget. Splitting $100 a day five ways starves the account in the same way fifty creatives do.
Under $100 a day there is no right number of creatives, because there is not enough budget to feed even one properly; the minimum budget post has that arithmetic. A budget out of line with the creative count is also one of the five things that kill a campaign that used to work. This is the third of five essentials in the video, between what you send back to Meta and finding where merchants drop off.