Data: June–September 2026 · 17 US ad accounts · $22.9M in Meta spend (June–August)
Part of our Meta Ads Statistics 2026 report.
The Short Answer
Two things are true about Meta ad accounts at scale, and they explain why creative volume matters so much:
- A few ads get most of the money. In the median account, the single biggest ad took about 10% of three months of spend, the top 10 took 41%, and just 16 ads made up half of all spend.
- Those winners don't last. Three months later, the top ads from June had kept a median of only 29% of their spend. By September, two out of three top-10 creatives were new since June.
In other words: most ads never get meaningful spend, the ones that do carry the account for a couple of months, and then they have to be replaced.
The Data
We pulled Meta data from 17 US ad accounts we have visibility into, each spending $100K or more from June through August 2026, for a combined $22.9M in spend. The accounts include consumer brands (food, supplements, pet, beauty and household) and apps and services (software, streaming, telecom, fintech and health).
- Spend concentration: each account's top 50 ads over June–August, compared to its total spend.
- Winner lifespan: each account's top 5 creatives in June, tracked month by month through September 26. We ran this on 7 of the largest accounts, which have enough ad volume to track, and matched the same creative even when it was relaunched as a new ad.
Brand names are anonymized.
Part 1: A Handful of Ads Get Most of the Spend
| Share of 3-month spend going to... | Median account | Range |
|---|---|---|
| The #1 ad | 10.5% | 2.7% – 23.9% |
| The top 5 ads | 29.4% | 9.2% – 57.8% |
| The top 10 ads | 40.7% | 14.3% – 75.2% |
| The top 50 ads | 77.3% | 36.5% – 99.7% |
- Half of all spend went to about 16 ads in the median account. In the most concentrated accounts, just 4 ads made up half the budget.
- The most diversified account spread its spend the widest. Its top 10 ads took only 14% of spend, and it took more than 50 ads to reach half. It was also the account that turned over its top creatives most completely by September (more on that below).
- Concentration is normal, not a warning sign. Meta is built to find the ads that work and push budget toward them. The goal isn't to spread spend evenly. It's to give Meta enough genuinely different ads that it keeps finding new winners.
We've written about this power law before in our breakdown of how Meta accounts spending $1M–$3M a month are structured. This data shows it holds across many more accounts.
Part 2: How Long Winning Ads Last
We took the top 5 creatives in June from 7 of the largest accounts and tracked them through September.
| After... | June's top 5 still in the top 10 | Spend retained (median) |
|---|---|---|
| 1 month (July) | 20 of 35 (57%) | 76% |
| 2 months (August) | 14 of 35 (40%) | 39% |
| 3 months (September) | 12 of 35 (34%) | 29% |
- Winners hold up for about a month, then fade fast. A month later, most of June's top creatives were still top-10 ads and still getting three-quarters of their spend. By month three, the typical winner had lost more than 70% of its spend.
- Twenty-one of the 35 June winners (60%) had dropped out of the top 50 entirely by September.
- Two out of three top ads turn over each quarter. Of September's top-10 creatives across these accounts, 46 of 70 weren't running at all in June. In the median account, only a third of September's spend went to creatives that existed in June.
Lifespan varies a lot by account
- One account had almost fully turned over. Only 2% of its September spend went to creatives from June. It launches new concepts constantly, and Meta keeps moving budget to the newest winners.
- One account was still leaning on old winners. 88% of its September spend went to creatives that were already running in June. That's a concentration risk: when those ads fatigue, it'll need replacements fast.
- Relaunching a winner can extend it. In one account, the top ad from June was duplicated into new ads over the summer, and one of those copies was the #1 ad again in September. A relaunch won't revive every tired ad, but it's worth testing for a true winner.
What This Means for Your Creative Pipeline
- Plan for a 2–3 month shelf life. Treat your best ads as temporary. If today's winners carry most of your spend, their replacements need to be in testing now, not after performance drops.
- Volume and variety both matter. With 16 ads making up half of spend and most winners fading within a quarter, you need a steady flow of new concepts to keep finding the next one. Variations of the same idea tend to fade together. See what creative diversity means on Meta.
- Watch concentration, not just performance. If one or two ads carry most of your spend, a single fatigued ad can drop the whole account. The more winners you have in rotation, the safer you are.
- Don't disrupt winners while you replace them. Launch new concepts in their own ad set so you're not pausing or moving what's working. We cover that in testing vs. scaling campaigns.
- Judge fatigue by results, not just time. Some winners lasted all four months. Retire an ad when its cost per result rises and spend shifts away from it, not on a fixed schedule.
FAQ
How long do Meta ads last before they fatigue?
In our data, top-performing ads held most of their spend for about a month. Two months in, they'd kept a median of 39% of their spend, and after three months just 29%. Plan for most winners to have a 2–3 month shelf life.
What is creative fatigue on Meta?
Creative fatigue is when an ad's performance declines because the same people have seen it too many times or it's stopped standing out. You'll usually see rising cost per result and falling click-through rates, and Meta shifts spend toward other ads.
How many ads get most of the spend in a Meta account?
Very few. In the median account in our data, the top ad took about 10% of three months of spend, the top 10 ads took 41%, and 16 ads made up half of all spend.
How often should you refresh Meta ad creative?
Continuously. Two out of three top ads in our data turned over within a quarter, so new concepts should be launching every week or every few weeks, before current winners fade.
Does relaunching a winning ad work?
Sometimes. In one account, a relaunched copy of June's top ad became the #1 ad again in September. It's worth testing for true winners, but it doesn't replace a steady supply of new concepts.
Where does this data come from?
17 US Meta ad accounts we have visibility into, each spending $100K or more from June to August 2026 ($22.9M combined). Winner lifespan was tracked June–September on 7 of the largest accounts. Brand names are anonymized. See also our Meta ads benchmarks and hook rate benchmarks.
Keep the Winners Coming
The accounts that scale treat creative as a pipeline, not a project. Biddyco is a creative agency for Meta ads for brands spending $100K–$5M+/month. Our in-house team makes UGC, motion, hybrid and static ads on a flat monthly retainer, so there's always a next winner in testing.
