How we'd structure WHOOP's Meta ads
Not affiliated with WHOOP. This analysis is based on publicly available information.
WHOOP does not publish acquisition costs or ad budgets. This plan is built on a €450 AOV and a €28,500 monthly budget, which is where a business at this scale plausibly sits. Change those inputs and the maths moves with them.
WHOOP sells a membership with the hardware included, and at around €450 to get in the door this is a considered purchase, not an impulse. The important thing about this account is that it is almost certainly already running. So the job here is not a fresh build. It is a rescue. The budget and the offer are both strong enough to work. What decides the result is whether the spend is pointed at one clean campaign or scattered across tiers and countries until no single ad set ever learns.
At €950 a day the account clears the fifty purchases a week Meta needs to exit learning, so you bid for the Purchase itself. No stepping down to Add to Cart.
The event closest to revenue. At this budget it fires often enough to teach delivery, so there is no reason to optimise for a softer correlate.
Why optimise for Purchase, not a softer event
The instinct at a €450 price point is to step the campaign down to a cheaper event, an Add to Cart or a lead, because purchases feel too rare to bid on. That instinct is wrong here, and the reason is the budget. At €950 a day against a €133 target cost per sale, the account clears roughly fifty purchases a week on its own.
Fifty is the number Meta needs to exit learning on an event. So you can bid straight for the Purchase and still get a stable, honest read. Optimise for a softer signal at this budget and you train the algorithm on a weaker stand-in for revenue, for no gain. Bid for the thing you actually want to happen.
The most you can pay per purchase and stay in profit. Every budget and scaling decision keys off this number.
Fifty purchases at €133 is €6,650, which lands at €950 a day. That is the minimum before the structure can be judged. Splitting it is the fastest way below the threshold.
Below this the campaign loses money on the first purchase. Read blended CPA against €133 and ROAS against 1.69x before deciding to scale.
Why €950 a day is a floor, not a target
The €133 target comes out of the unit economics. The daily number comes out of the learning phase. Fifty purchases at €133 is €6,650, and that is the minimum spend before you can fairly judge whether the structure works. Across a week it lands at €950 a day, and the €28,500 monthly budget already covers it.
The trap is not the size of the budget. The trap is slicing it. Spread €950 across four campaigns and each one drops below the fifty-purchase line, so every ad set sits in permanent learning and the cost per sale never settles. One pool of spend learns. Four puddles do not.
US, UK, and EU-core each get one broad campaign, one or two ad sets, three to five ads. This is a restructure of a fragmented account, not a fresh build.
No interest layering. Split by country, where delivery and currency genuinely differ. Never split by audience segment within a country.
Merge campaigns split by tier, country, or interest into one broad CBO campaign per market. Everything else waits until the purchase signal is pooled.
The call that matters: consolidate
This is the single most important decision on the page, and it has nothing to do with creative. A brand at this scale, running in several countries, almost always ends up with a fragmented account. One campaign per membership tier. An ad set for athletes, another for biohackers, a lookalike bolted on. Each split feels like control. In practice it starves every ad set of the purchase volume it needs to learn.
The fix is boring and it works. One broad campaign per major market, US, UK, and EU-core, with campaign budget optimisation on and one or two broad ad sets underneath. Split by geography, because delivery and currency really do differ by country. Do not split by audience, because Meta finds the buyer faster than a hand-built interest stack does. Consolidate first. Judge second.
1) Fragmenting budget across membership tiers or countries into too many small campaigns. This is the most likely cause of the current underperformance, and it pushes ad sets below the fifty-purchase threshold.
2) Layering interest targeting, fitness, biohacking, athletes, on top of broad. This is a wide B2C buyer base, and in my experience broad delivery finds these buyers more cheaply than a hand-picked interest stack once the pixel has a real purchase signal to learn from.
3) Editing live campaigns to chase daily swings. Any targeting, creative, or bid change resets learning, and at €950 a day you cannot afford repeated resets against a €133 target.
The offer carries this, so test the offer
The offer is the strongest asset here, so the first creative should say it plainly. WHOOP's own page leads with numbers: 91 more minutes of activity a week, 2.3 more hours of sleep, over 10% higher HRV. The ad should lead with those too, not soften them into a lifestyle shot. Run that data-forward static against the free-trial video, the same audience behind both, and let spend pick the winner.
Then get the measurement honest before you scale. Move the Purchase event server side with the Conversions API, or you will underreport and mislead the bidder. Leave it alone for a full seven days. Read the winner against the €133 target and the 1.69x break-even, and if it is clearing, lift budget in twenty percent steps rather than rebuilding the account again.
Move into the new structure and do not touch it. Expect a volatile daily cost while the algorithm learns. That is normal, not failure.
Check purchases per ad set against the fifty mark and the cost trend against €133. This is information, not yet a decision.
Decide on blended CPA versus €133 and ROAS versus 1.69x. If it is consistently under target, raise budget in twenty percent steps rather than restructuring again.