The headline acquisition campaign ran at a 1.35x return on ad spend. On a dashboard that looks like a campaign to throttle. It was the opposite. Behind it sat a 33% returning customer rate and around $202K a year in subscription revenue, and the store grew 437%.
The brand sells ingestible botanical wellness products. Nectar, gummies and a facial mist, built around a single hero ingredient, sold direct in the US with a subscription option.
Its Meta acquisition looked mediocre on the only number most people check. The primary always-on campaign carried a first-order return on ad spend of 1.35. Several supporting campaigns sat between 0.76 and 1.16. Judged as a standalone media buy, that is roughly break-even before cost of goods, and it is the kind of number that gets a campaign paused.
The store was not break-even. Over the twelve months to November 2025 it did $1,288,005 in total sales, up 437% year on year, across 14,796 orders, up 413%, at an average order value of $81.67.
A third of customers came back. Subscriptions added roughly $202,000 across the year on top of one-time orders. The two newest products grew 560% and 680% year on year, largely to people who had already bought the hero product.
First-order ROAS is a measure of the first order. It is not a measure of the customer. On a subscription or repeat-purchase brand, holding a 1.35x on acquisition is a decision to buy a customer worth several times the first basket. An agency optimising to a 2x first-purchase target would have throttled the exact campaign that built the year.
The account was managed to cost per acquisition rather than to first-order ROAS. A capped-CPA prospecting campaign carried the volume at $58.70 per purchase and ran continuously rather than being paused every time in-platform return dipped below breakeven.
Around it sat product-focus prospecting for each individual SKU, so the two newer products got their own demand rather than relying on cross-sell, and a re-engagement layer for people who had bought once and not subscribed.
The discipline was the boring part: hold the CPA ceiling, do not react to a ROAS number that only describes day one, and let the subscription and repeat revenue land where it lands.
Total sales of $1,288,005 over the twelve months, up 437%. Orders up 413%. Average order value up 8%, so the growth was not bought by discounting into a smaller basket. Online store revenue specifically grew 581%.
In the most recent 30 days of the period, the primary campaign was still holding, at $63.33 per purchase and a 1.48x first-order ROAS on $16,657 of spend.
Figures are from the brand's Shopify analytics and Meta Ads Manager for the twelve months to November 2025. Multiple partners have worked in this ad account over its lifetime.
A hero ingestible plus two follow-on products, sold direct with a subscription option. The two newer SKUs grew 560% and 680% year on year, largely to customers who had already bought the hero.
On its own, no. In a business where a third of customers come back and subscriptions add six figures a year, it can be very good. The number that matters is what a customer is worth over their life against what they cost to acquire, not what the first order returned.
You need three things before you can defend it: a repeat rate you can measure, a subscription or reorder revenue line you can see separately, and a cost per acquisition ceiling you actually hold. Without those, a low ROAS is just a low ROAS.
No. If customers buy once and never return, first-order ROAS is close to the whole picture and you should treat it that way. This approach only applies where repeat behaviour is real and measurable.
We'll look at repeat rate and subscription revenue next to acquisition cost and tell you what you can actually afford to pay.
Get a free audit