MHI Media fashion brand case study: Meta ads (anonymised)
This anonymised MHI Media case study follows a US fashion brand on Meta (Facebook and Instagram) ads. Between month 1 and month 11 of working with us, monthly Meta spend grew from about $18k to about $41k, roughly 2.3 times. Purchases tripled, CPA fell about 25% and ROAS rose about a third. From month 4 onwards, ads we made carried more than 90% of the account's spend.
We've left out the brand's name, products and people, and anything else that could point to it. Figures are rounded and come from the brand's Meta ad account as Meta reports them. Month 1 is the first full calendar month we worked together.
The starting point
This is a smaller account. In month 1 the brand spent about $18k on Meta, and in the three months before that it had spent between about $18k and $29k a month.
The results section compares month 11 with month 1, and with the average of those three earlier months.
Our first ads went live halfway through month 1. Until then, everything in the account came from the brand's existing creative. Over the next three months our ads replaced it. The aim was ads that could hold CPA as spend went up, at a budget where every test has to earn its place.
What we changed
By month 4 our ads carried all of the account's spend, and they have carried more than 90% in every month since.
About three-quarters of the spend on our ads went to new concepts. The other quarter went to new versions of ads that had already worked, mostly with a new hook or a new edit. About 70% of the spend on our ads went to video. Static images took the rest.
New ads went live almost every week. At this budget an account can't test everything at once, so each week's batch had a clear job: test a new idea, or get more out of an ad that was already working. Hooks were tested in versions, with the same body and a different opening, so a good idea didn't get thrown out because of a weak first few seconds.
Results
| Measure | Month 1 to month 11 | Month 11 against the three months before month 1 |
|---|---|---|
| Monthly Meta spend | About $18k to about $41k (about 2.3x) | About 1.65x |
| Purchases a month | About 3x | About 2x |
| CPA | Down about 25% | Down about 19% |
| ROAS | Up about a third | Up about 19% |
| Average order value | About flat | Down about 4% |
| Share of spend on ads we made | About 14% to about 92% | None to about 92% |
Source: the brand's Meta ad account, as reported by Meta.
CPA was lower and ROAS higher than month 1 in every month from month 2 onwards. The best CPA came in month 5, and spend peaked in month 10 at a little over $50k.
Against the average of the three months before month 1, month 11 shows about 1.65 times the spend, a CPA about 19% lower and a ROAS about 19% higher.
What other fashion brands can take from this
If you run a fashion brand spending under $50k a month on Meta, this account is a useful reference point.
- Small budgets still need a steady flow of new ads. New ads went live almost every week here, and CPA stayed below month 1 as spend grew. See how to slow creative fatigue.
- Spend part of the budget on new versions of winners. About a quarter of the spend on our ads went to iterations. A new hook on an ad that already works is cheaper to test than a new concept.
- Test in a way your budget can read. Our creative testing framework covers how to set this up at different spend levels.
- Check your numbers against your category. The fashion Meta ads benchmarks show typical CPA and ROAS.
For bigger fashion accounts, read the Ripley Rader and BeyondNine case studies, or see our fashion brands page. To talk about your own account, book a strategy call.
FAQ
Can a small fashion brand grow on Meta ads without CPA going up?
Yes, if the creative keeps up. In this account monthly Meta spend grew from about $18k to about $41k while CPA fell about 25% against month 1, and about 19% against the three months before we started. At smaller budgets each new ad matters more, because there are fewer ads in the account to carry the spend.
How often should a fashion brand launch new Meta ads?
Often enough that a working ad is ready before the current ones tire. In this account new ads went live almost every week, with about three-quarters of the spend on our ads going to new concepts and a quarter to new versions of ads that had already worked. The right pace depends on your spend, but weekly is a good starting point.
Why does this case study show two comparisons?
So you can see the result against our first month and against the period before we started: month 1 to month 11, and month 11 against the average of the three months before. Both show more spend at a lower CPA and a higher ROAS.
Sources
- The brand's Meta ad account (Facebook and Instagram), the three months before month 1, and months 1 to 11, figures as reported by Meta, rounded