Advantage+ allocates budget between distinct creatives, and a distinct creative is one with its own opening shot and its own script. Everything else you loaded, the four aspect ratios, the placement crops, the six primary text variants, is packaging around a creative the system already has. Most accounts we look at are running four to eight distinct creatives while their reporting shows twenty ads, which is why the campaign keeps concentrating spend into the same two winners.
The fix is not more assets. It is knowing which number you are actually feeding the optimiser, then raising that number rather than the file count.
What Advantage+ treats as one creative
The system makes two separate decisions and it helps to keep them apart. The first is which creative to show a given person. The second is how to package that creative for the placement it lands in. Placement asset customisation, automatic cropping and the ratio set all sit in the second decision, and Meta will make them for you.
So the operational definition of distinct is narrow: two creatives are distinct if a viewer who saw one and then the other would experience two different ads in the first two seconds. Same first frame plus same first line of narration means one creative, regardless of how many ad IDs it has.
| What the interface shows | What the optimiser gets | What your report counts |
|---|---|---|
| 1 video, exported at 9:16, 4:5, 1:1 and 16:9 | 1 creative | 4 ads |
| 1 video, placement asset customisation on | 1 creative | 1 ad, several deliveries |
| 1 video, 6 primary texts under DCO | 1 video creative, 6 copy combinations | 1 ad, combination-level breakdown |
| 6 videos, each duplicated into 3 ad sets | 6 creatives | 18 ads |
| 4 videos, same footage, same hook, different shot order | 1 to 2 creatives | 4 ads |
| 4 videos, different footage, different hooks, different scripts | 4 creatives | 4 ads |
The third row is the one that surprises people. A dynamic creative optimisation setup with one video and six texts produces six combinations, and the combination report makes that look like six tests. The video is constant across all six, so the biggest lever in the ad is untested. Combination count and creative count are different quantities and Ads Manager encourages you to conflate them.
How to audit an existing Advantage+ campaign
Ad counts overstate variety, so measure spend concentration instead. This gives you an effective creative count, which is the number of creatives the campaign behaved as if it had.
- Pull the last 14 days of spend, broken out by ad, for one Advantage+ campaign.
- Collapse the rows into distinct creatives using the definition above. Sum spend within each group.
- Convert each group's spend to a share of campaign spend.
- Square each share and add the squares together.
- Effective creative count = 1 divided by that sum.
effective creative count = 1 / (s1² + s2² + ... + sn²)
Worked example. A campaign shows 24 ads. Collapsing gives 5 distinct creatives with spend shares of 45, 25, 15, 10 and 5 percent. The squares are 0.2025, 0.0625, 0.0225, 0.0100 and 0.0025, summing to 0.30. Divide 1 by 0.30 and the effective creative count is 3.3.
That campaign had 24 rows in its report, 5 real creatives, and behaved like 3.3. Substitute your own shares. If all five had drawn an even 20 percent, the sum of squares would be 0.20 and the effective count would be exactly 5, which is the most the set could give.
Delivery concentration is not a bug, it is the optimiser doing its job with what it has. An effective count well below your distinct count means the extra creatives are not competitive, which is a creative signal to act on rather than a targeting problem.
Why the range is 8 to 20 and not 40
Our 2026 ad fatigue benchmark, which synthesises published platform and agency figures rather than measuring anything ourselves, puts 8 to 20 live variations as what a typical active campaign needs in rotation. The range is worth deriving rather than accepting, because the derivation tells you where in it you belong.
A live rotation is a queue. Creative enters when you ship it and leaves when it decays, so the steady-state size is your ship rate multiplied by how long a creative survives:
live set size = new creatives per week × creative lifespan in weeks
The benchmark puts most creative as effectively dead within three weeks, with week three sitting 45 to 70 percent below the launch baseline. It also associates shipping 15 to 50 variants a month with 3 to 5 times longer campaign lifespan than quarterly refreshing. Fifteen a month is about 3.5 a week, fifty is about 11.6.
| Ship rate per week | Lifespan | Steady-state live set |
|---|---|---|
| 2 | 3 weeks | 6 |
| 3.5 | 3 weeks | 10.5 |
| 5 | 3 weeks | 15 |
| 7 | 3 weeks | 21 |
| 5 | 4 weeks (B2B SaaS, 28 days to a 40% decline) | 20 |
| 5 | 1.3 weeks (food and beverage, 9 days) | 6.5 |
The lower bound comes from frequency. Meta prospecting starts declining at a weekly frequency of 2.5, and Meta internal research cited in the benchmark reports CTR dropping 45 percent after a fourth exposure to the same creative. A rotation of four with concentrated delivery pushes its top creative past that inside a week. Eight is roughly where the top creative's share of impressions stays survivable.
The upper bound is budget, not fatigue. Every creative you add divides the same impressions further, and past 20 most accounts cannot give each one enough impressions to tell a winner from noise. If your effective creative count is stuck at 3 while your distinct count is 15, adding a sixteenth is not the move.
Placement exports versus creatives, in practice
Export all four ratios. It costs nothing and it stops Meta cropping a caption off the bottom of a Reel. Just do not book those exports as variety: keep two columns in your creative tracker, one for concepts shipped and one for files delivered. A team shipping 5 concepts a week at four ratios each delivers 20 files, and reporting 20 as the creative number is how a campaign ends up with an effective count of 3 and nobody noticing for a month.
Raising the number that matters
Genyad is our product, so read this as disclosure. It takes a library of video you already own, transcribes and tags every clip on upload, then builds each variation as a fresh script with its own shot selection, voiceover and caption set drawn from that library rather than a re-cut of one timeline. That is the reason its output moves the distinct count instead of the file count, and it is the whole argument for our Meta ad creative generator workflow. See the definition of an ad variation if you want the distinction stated on its own terms.
One variation is 1 credit, all four ratios included, so a 5-a-week cadence runs 5 credits a week. Editing and re-exporting are free. Scripts are written natively in English, German, French, Spanish, Italian or Hindi rather than translated, which matters if your Advantage+ campaign spans markets: a German script written as German is a distinct creative, a subtitled English one is a packaging change.
What we do not do: no AI avatars or synthetic presenters, no static banners, no product-URL import, no product-feed or CSV-driven template rendering, no predicted performance scores, and no direct publishing into Meta, so you download the exports and upload them yourself. If your bottleneck is rendering thousands of catalogue permutations from a feed, a template rendering tool is the right category and we are not in it.
Frequently asked questions
Does Advantage+ need fewer creatives because it optimises automatically?
No, it needs more usable ones. Automatic optimisation decides how to allocate between the creatives you supply, so its ceiling is set by the pool, and a pool of three cannot be optimised into a pool of twelve. Our 2026 fatigue benchmark puts a typical active campaign at 8 to 20 live variations, and that figure does not fall because the campaign type changed.
Do the four aspect ratios count as four creative variations?
No. Four ratios of one video are four packages of one creative, and Advantage+ would have handled the resizing itself through placement asset customisation. Export them anyway, because a proper 9:16 beats an auto-crop for caption safety, but count them as one creative when you plan rotation size.
What is a good effective creative count?
Aim for it to sit within about 60 percent of your distinct creative count. Five distinct creatives behaving like 3.3 is normal for a campaign with one clear winner. Five behaving like 1.5 means one ad is taking nearly all the budget, which is a signal to ship replacements rather than to add more of the same concept.
Should I put all my creatives in one Advantage+ campaign or split them?
Consolidate unless you have a reporting reason to split. Splitting the same creatives across several campaigns divides the impressions each one gets, which makes every read noisier and slows the optimiser's learning without adding any variety. Splitting earns its keep when budgets are ring-fenced by market or product line, not as a way to test creative.