Blog Playbooks · The Benchmark Playbooks · Part 3
Where your Meta ad money really goes, and when to refresh a winner
Nearly half of all the money in 18,920 Meta ads went to the 6.2% that became winners. Here is how to manage that concentration, spot a tiring winner and replace it before it fades.
If you run Meta ads for a local business, a few of your ads are doing most of the work, whether you planned it that way or not. We have run over 18,920 Meta ads and graded each one in Creative Benchmarks. The 6.2% that became winners took 49% of all the money. Inside a typical account, the top 6% of ads took 42% of its spend.
That concentration is not a problem to fix. It is how Meta's delivery works, and it is usually good for you. The risk is what happens when one of those few ads gets tired, because a large slice of your budget is riding on it. This playbook covers how to see your own concentration, how to tell a tiring winner from a normal bad week, and how to refresh without starting from zero.
What the benchmark shows
Source: Creative Benchmarks
A winner in Creative Benchmarks is an ad that spent at least 10 times its own account's median ad, and at least $500. By definition winners take a lot of spend. What the numbers add is the scale: roughly one ad in sixteen takes about half the money.
Put it in hypothetical terms. Say you spend $1,500 a month and have 50 ads that have run this year. The top 6% is three ads. If your account looks like the typical one, those three took about 42% of the spend, or about $630 of each $1,500. The other 47 shared the remaining $870.
So when one of those three ads starts to slip, it is not one ad in fifty slipping. It is one of three ads holding up 42% of your budget.
When a winner slips, it is not one ad in fifty slipping. It is a large share of your budget.
Spend is not the same as results
Before you protect your winners, check that they deserve it. Spend tells you which ads Meta chose. It does not tell you they bring the cheapest leads. In our data, 47 in 100 winners brought leads cheaper than their own account's average. The other 53 did not.
Source: Creative Benchmarks, benchmark.sneeze.it
So a winner can be a winner on spend and still be an expensive source of leads. That is not a reason to pause it on sight; a winner with an average cost per lead may still be bringing most of your volume. It is a reason to know which kind of winner you have. Grade every ad twice covers that check in detail.
How to tell a winner is tiring
There is no single number that marks fatigue, and we will not give you one we cannot back. What works is comparing the winner with itself.
A winner that is tiring usually shows a pattern over several weeks, not one bad day: its cost per lead creeps up week over week, while it keeps taking a similar share of spend. One expensive week can be noise. Three weeks in a row moving the same way is a signal.
Also check whether anything else changed. A new offer, a price change, a holiday or a change to your website can all move cost per lead across every ad at once. If all your ads got worse together, the problem is probably not the winner.
The playbook: manage the few ads that carry you
- Find your top ads
Sort every ad in the account by lifetime spend and mark the top 6% (or your top three to five, in a smaller account). Work out what share of total spend they took.
- Grade each one on cost per lead
Compare each top ad's cost per lead with your account average. Mark which ones are cheaper and which are not.
- Track them weekly
Each week, log spend and cost per lead for each top ad. You are looking for a steady rise over several weeks, not a single bad week.
- Build the replacement before you need it
For each winner, make two or three new ads that keep what worked (the look, the opening, the offer) and change one thing. Launch them while the winner is still running.
- Do not pause the winner first
Let the new ads compete alongside it. If one starts taking spend and holds a good cost per lead, Meta will shift money toward it on its own. Pause the old winner only once a replacement is carrying real spend, or if its cost per lead has clearly passed your account average and stayed there.
- Keep launching new ideas too
Variations of a winner extend its life. New ideas find the next winner. You need both, at a pace your budget can carry. See how many new ads a week.
Why concentration is fine, and dependence is not
It is normal for a handful of ads to carry an account. The benchmark says so plainly, and fighting it by forcing even spend across ads usually just pushes money to the ads Meta had already learned were weaker.
The danger is having one winner and nothing behind it. When that ad tires, results drop, and the next winner takes weeks to find because nothing new was in the pipeline. The fix is a steady supply of new ads and variations, launched before you need them, plus a weekly look at the few ads that matter most. Most new ads will barely spend (two in three never reach $100, as most ads barely run shows), so the pipeline has to be wider than you would think.
