Run three to five creatives per audience at once, not one and not thirty. One ad gives you no way to tell whether a result came from the creative, the offer, or the week. Thirty ads on a small budget give each creative too little delivery to produce a readable number. Three to five is where a real spread appears.
The brands that advertise every day settle this question very differently than most small advertisers expect, and the gap between the two is worth seeing before you pick your own number.
How many ads do big brands run at the same time?
The median brand runs about 170 ads concurrently. Sixty-two brands run more than 500 at once. Fifty-three run fewer than 50. These figures come from the Hello Radio archive, a live archive of real ads from more than 300 brands refreshed every morning, measured in August 2026.
Not 170 ads over the course of a year. 170 at the same time.
Some specific counts, all measured the same month: Sephora runs 979 ads concurrently. Paramount Plus runs 921. HOKA runs 868. Starbucks runs 863. Petco runs 910. These are not scattered experiments. They are the normal operating state of a company that treats advertising as a permanent function rather than a seasonal campaign.
The point is not that you should run 170 ads. You should not. A brand running 900 ads is running many products, many geographies, and many audiences at once, and it has the budget to give every one of them enough delivery to matter. The point is that the sensible ceiling is much higher than most small advertisers imagine, and the floor is higher too.
Why one ad is always the wrong number
If you run a single ad and it performs badly, you have learned almost nothing. You do not know whether the image was wrong, the headline was wrong, the offer was wrong, or the audience was wrong. You know only that this exact combination did not work this week.
If you run a single ad and it performs well, you have learned slightly more, but you still have no idea how much better it could have been. You have set a benchmark of one, against nothing.
Two ads is better but unstable. When one wins, the margin usually sits inside the noise. Three to five is where you start to see a real spread: one clearly ahead, one clearly behind, the rest bunched in the middle. That spread is the information you are actually paying for.
The budget math that sets your real ceiling
Every ad you add divides the same budget into smaller pieces, and below a certain daily delivery, results stop being trustworthy. A creative getting a handful of impressions a day produces numbers that look like performance but are mostly randomness.
So your ad count is not a preference. It is your budget divided by the minimum spend each creative needs to produce a readable result.
That is the honest answer to the instinct to launch ten creatives on a small budget to see what sticks. Ten creatives on a small budget produce ten unreadable results, and the winner you pick is usually just the one that got lucky. See how to test Facebook ads before spending big money for the sequencing that avoids this.
How to set the number for your account
| Stage | Creatives live at once | What you are trying to learn |
|---|---|---|
| Starting out | 3 | Which angle gets any traction at all |
| Something works | 3 to 5 | Whether a challenger can beat the winner |
| Scaling | 5 to 10 | Which angles hold up as spend increases |
| Real volume | 10 or more | Which products and audiences deserve more creative |
If you are starting out. Three creatives, one audience, one clear offer. Different angles, not different crops of the same photo. Give them enough days and enough budget to separate.
Once something works. Keep the winner running and add two or three challengers against it. Do not swap the winner out because you are bored of it. The entire reason to find a winner is to keep spending behind it.
As you scale. Add creatives before you add complexity. More angles against a proven audience is almost always safer than the same creative against speculative new audiences.
When you have real volume. This is where your account starts to resemble the brands in the archive: many creatives across many products and audiences, with a steady pipeline replacing whatever fatigues.
The pattern worth copying from big advertisers
Large advertisers do not run hundreds of ads because they cannot decide. They run hundreds because they treat creative as inventory. There is always a set producing, a set being tested, and a set on the way out. Nothing launches into an empty account, and nothing that works gets turned off for being old.
That structure is the real lesson, and it scales down. Even at three creatives you can run it: one proven, one challenger, one new angle. The ratio matters more than the number.
Ad count is also the wrong metric to obsess over on its own. What matters is whether each running ad has enough delivery to tell you something, and whether you have a next creative ready when the current one fades. On that second point, see how often you should refresh ad creative.
Where the angles come from
The hard part of running more ads is not the ad manager. It is having enough genuinely distinct ideas to fill the slots without producing five near-identical creatives that all fail the same way.
That is what a live archive of working ads is for. Hello Radio keeps a running archive of ads from more than 300 brands, refreshed every morning, so the angles big advertisers are actually paying to run are in front of you when you sit down to write. Browse it in the library, or connect it to Claude through the Hello Radio connection and ask for the patterns in your category.
Common questions
How many ads should be in one ad set?
Three to five creatives in a single ad set is the practical range for most advertisers. Fewer than three and you cannot tell a winner from noise. More than five on a modest budget splits delivery so thinly that none of the results are readable.
Does running more ads increase costs?
Not directly. You set the budget, and more ads divide that same budget rather than adding to it. The indirect cost is that too many creatives on too little budget produce unreliable data, which leads to bad decisions.
Should I turn off losing ads immediately?
No. Give every creative enough delivery to produce a stable number first. Turning off an ad after a few impressions usually means killing something on randomness rather than on performance. See how long it takes for Facebook ads to work.
Can I run the same ad to multiple audiences?
Yes, and it is usually a better use of a proven creative than inventing a new one. Just keep the audiences separate enough that you can read the difference, and exclude overlapping groups so you are not paying twice to reach the same person.