Case study Results / E-commerce · Subscriptions
£123 to win a customer worth £309.
A supplements brand on a subscription model, where the first order is never the point. Before we could set a single bid we had to answer a question nobody had asked: what is a subscriber actually worth? That answer, £309, is what allowed the account to spend £123 to win one and scale hard rather than optimise for a cheap first sale.
430 subscriptions · £123 each · £309 modelled value · six months
One of thirteen case studies, published from more than a hundred accounts since 2021. We publish the ones we can put a source against. See all thirteen →
01 The starting point
Plenty of activity. No arithmetic.
The account arrived fragmented, with tracking that could not be trusted and a budget small enough to be mostly waste. None of that was the real problem. The real problem was that nobody in the business could say what a new subscriber was worth, so every decision about what to pay for one was a guess dressed as a strategy.
Get that number wrong in a subscription business and you either starve a profitable account or scale a loss. Most brands never work it out at all.
02 The arithmetic
Where the £309 came from.
Three numbers from the client’s own books: an average order of £65, a typical subscriber life of around six months, and a quarter of new subscribers not ordering again after the first month. That is enough to value a customer, and valuing a customer is enough to set a target.
The client then made the call that mattered: scale over efficiency. A £200 target cost per acquisition left margin intact while letting the account grow quickly. It came in at £123.
03 What we did
What we did
Worked out the unit economics first. Average order value, churn, repeat rate, subscriber life. Until those exist, a target cost per acquisition is invented. Once they exist, it is arithmetic, and the client can sign it off in a minute because they recognise their own numbers.
Asked the client to choose: scale or efficiency. Those are different accounts. They wanted scale, so the target was set to allow it while keeping margin, rather than defaulting to the cheapest possible sale and quietly capping the business.
Rebuilt tracking with server-side tagging. Subscription businesses lose conversions to browser restrictions and ad blockers, and a subscription conversion that goes missing is not one lost sale, it is six months of them. Server-side tagging closed that gap.
Built a funnel with a job at each stage. Demand Gen and video to create awareness, Search, Shopping, dynamic search ads and Performance Max to catch intent, dynamic remarketing and brand to close. Each stage measured on what it is actually for.
Tested the structure rather than assuming it. Performance Max against feed-only Performance Max, feed-only against standard Shopping, high-intent against low-intent non-brand search, dynamic search ads pointed at blog pages against product pages. The account settled where the evidence pointed.
Scaled when the numbers said to, not before. Spend moved from £7,000 a month to £15,000 to £26,000 as the cost per subscription held under target. More went through the account in four months than in the previous two years.
Do you know what a customer is worth?
If the answer is not a number, the account is guessing. The free audit starts with the arithmetic, then looks at the ads. Walked through with you, live on a call or as a recorded video.
Free audit · live or recorded · no obligation