Blog Playbooks · The Benchmark Playbooks · Part 1
How many new Meta ads should you launch a week?
Accounts that launched four or more new ads a week found about one winner a month. Accounts under one a week found about one a quarter. Here is how to pick your pace.
Most local businesses running Meta ads ask the wrong question about creative. They ask "which ad is best?" when the better question is "how many ads do we give Meta to choose from?" We have run over 18,920 Meta ads and graded every one of them in Creative Benchmarks. Only 6.2% became winners. If one ad in sixteen is going to carry the account, the number of ads you launch matters as much as how good any single one is.
This playbook covers one finding: how often accounts found a winner, sorted by how many new ads they launched each week.
What the benchmark shows
First, the definition. In Creative Benchmarks a winner is an ad that spent at least 10 times its own account's median ad, and at least $500. It is measured against its own account, so a small account and a large account are judged on the same scale. Meta only keeps spending on an ad when it keeps finding people to show it to, so a winner is an ad Meta chose to scale.
Then we looked at the 97 accounts with at least 50 leads and asked how many winners each found per month, grouped by how many new ads it launched per week. These are medians.
Source: Creative Benchmarks, benchmark.sneeze.it (median across 97 accounts with 50+ leads)
The typical account launched 1.6 new ads a week, which puts it in the second group: about 0.43 winners a month, or about one every two to three months. Accounts launching four or more a week found about one a month.
Run the arithmetic over a year and the gap gets plain. At 0.37 a month you would expect between four and five winners in twelve months. At 1.00 a month you would expect twelve.
If one ad in sixteen is going to carry the account, the number of ads you launch matters as much as how good any one of them is.
What it means, and what it does not
Be careful with this chart. It shows that accounts which launched more ads found more winners. It does not prove that launching more ads caused the extra winners. Accounts that launch often may also have bigger budgets, more locations or a team that makes creative faster, and any of those could help.
What we can say is that the pattern runs the right way at every step, and it fits how Meta works. A winner is picked by Meta's delivery, not by you. Each new ad is another draw. With 6.2% of ads becoming winners overall, an account that launches one ad a week is drawing about four times a month, and an account that launches four a week is drawing about seventeen.
There is also a ceiling the chart does not show. Each ad needs some spend to get a fair look. If your budget is small, ten new ads a week will split it so thin that none of them gets a real chance. Cadence has to fit the budget.
A quick hypothetical to make the winner line concrete. Say your median ad spent $40 over its life. Ten times that is $400, which is under the $500 floor, so your winner line is $500. If your median ad spent $80, the line is $800.
The playbook: set a launch rhythm you can keep
- Measure where you are
Count new ads launched per week over the last 30 days, and winners over the last six months, using the method above. Write both numbers down before changing anything.
- Move up one band, not three
If you launch under one a week, aim for one to two. If you launch one to two, aim for two to four. A pace you hold for three months beats a burst you drop after two weeks.
- Check the budget can carry it
Divide your monthly budget by the number of new ads you plan to launch that month. Say you spend $1,500 a month and plan 12 new ads: that is $125 of room per new ad, before your existing winners take their share. If the number looks too thin to tell you anything, launch fewer and make each one count.
- Make variety, not copies
Five versions of the same ad with a new headline are close to one draw. Change the look, the opening or the production style between ads so each launch is a real test. The cost per lead by kind of ad is a good menu to pick from.
- Batch the work
Make a month of ads in one sitting and schedule them to start across the weeks. Cadence fails when creative is made one ad at a time, on the day it is due.
- Leave the old ads alone
Launching new ads does not mean pausing the ones that are working. Let Meta keep spending on current winners while the new ones get their look.
- Review monthly
Once a month, count new ads launched and winners found. If you held the pace for a quarter and winners did not move, look at variety and budget before you add more volume.
How this fits with the rest
Cadence answers how many ads to make. It does not tell you how to judge them once they are live, or what to do when a winner starts to tire. Read most ads barely run for how to read an ad in its first week without cutting a winner early, and where the money goes for what to do once a few ads are carrying the account.
