Why do a few ads end up with most of your ad spend?
Ad creative follows a power law. Your top 1 to 2% of ads should be doing about 50% of total spend. Expect roughly 85% of creative to fail, and treat a blended hit rate of 10 to 15% as healthy.
- Variations · 20% hit rateThese are new versions of what already works, drawn from winners, top angles and proven hooks.
- Swings · 10% hit rateThese come from outside the account entirely, and they're the ads that find the outlier.
A few ads take most of the spend.
Most people plan creative as if results are normally distributed. Make twenty ads and you expect a few good ones, a few bad ones and a cluster of average ones carrying the account. That isn't what happens.
A very small number of ads absorb an enormous share of spend and return, and almost everything else does nothing. The distribution has a long tail, and the money is in the tail. Your job is to make enough attempts that an outlier shows up, then get out of its way when it does.
That changes what a batch of creative is for. A batch isn't a set of ads you expect to work. It's a set of lottery tickets bought at a known price, with the payout concentrated in one or two of them.
The thresholds I work to.
- 01The top 1 to 2% should take half the spendIf spend is spread evenly across your library, the algorithm isn't being allowed to concentrate, and you're funding mediocrity.
- 02A 10 to 15% blended hit rate is healthyThat's the zone to aim for.
- 03Above 20% means you're testing too safelyA high hit rate says your ideas are close to what already works, so you aren't buying enough variance to find the next outlier.
- 04Around 85% of creative failsAccept it up front and it stops feeling like a series of disasters.
Split production in half.
This is for brands spending $250K+ a month.
Half the batch is variations, new versions built from what your data already says works, drawn from winners, top angles and proven hooks. Aim for a 20% hit rate here. These are your safe bets, and they keep the account fed.
The other half is swings, ideas from outside the account entirely. They come from competitor concepts, customer calls, brainstorms, organic content, formats that feel risky and producers with a different style. Aim for a 10% hit rate. These are the ads that find the outlier.
Blended, that lands you in the 10 to 15% zone. If your blended rate climbs above 20%, you're being too safe, so shift budget from variations to swings.
A 30% hit rate was the problem.
A founder came in spending $150K a month on Meta at a $110 CPA. They were proud of a 30% hit rate and thought that was the win. It was the problem. A hit rate that high meant almost every ad was a minor variation of the last one, so nothing ever broke out.
The first phase was media buying alone, letting winners take budget instead of capping them. That cut CPA by 48%. The second phase was creative, once the swings started landing, and it took another 50% off. The total reduction was 74%.
Spend tripled to $500K a month, and their top 1 to 2% of ads ended up doing more than half of it. The company later sold for over $100 million.
None of this needed a higher hit rate. The hit rate went down. What changed is that when an outlier appeared, it was allowed to take budget instead of being capped to avoid putting all the eggs in one basket.
When it doesn't hold.
There are five cases where these numbers don't apply.
- 01Below roughly $250K a monthThe split is less useful. You aren't running enough ads for the distribution to show, and each test takes longer to read.
- 02A hit needs a definitionFix the threshold first, whether that's spend absorbed, CPA against target or a significance test, and hold it constant. Otherwise the number moves because your definition moved.
- 03The 1 to 2% is an outcomeIt describes how spend should end up distributed. Don't force it by pushing budget into two ads you like.
- 04Capping winners breaks itThe power law depends on the algorithm being allowed to consolidate. Capping winners to spread risk, or splitting budget evenly across ad sets, flattens the distribution and the model stops applying.
- 05New accounts and new offers behave differentlyBefore there's any performance history, everything is a swing, and the variations half has nothing to draw on yet.
What to do with it.
Two numbers tell you whether this is working in your account. One is the share of spend going to your top 1 to 2% of ads. The other is your blended hit rate, against a fixed definition of a hit.
Plot the account on a log-spend scatter and you'll see the concentration right away. Either a few ads have pulled away, or everything is bunched together and nothing has been allowed to win.
Plot every ad on spend against CPA, with confidence bands.
Free, and it runs in your browser. Nothing gets uploaded.
Get the weekly email.
One tactic a week, to 3,475 operators.