A cost ceiling that holds at volume
Leadership ran the cohort math and drew a line: a first deposit has to cost under $180, or the cohort never pays back. The job was to buy more users without crossing it. A buying system built on account limits, an approach backlog and constant rotation now delivers a steady daily flow — 828 first deposits in a month, cost held between $150 and $170.
The company
A product company of around a hundred people. Growth runs on paid traffic, so the cost of a new paying user is not a marketing detail — it decides whether each monthly cohort of users pays back or loses money.
The problem
Leadership had done the math and drawn a hard line.
- Looking at past months, they worked out the payback point: keep the cost per first deposit under $180, or the cohort never earns back what it cost to acquire.
- The goal was to buy *more* users while staying under that line — and those two pull against each other. The usual way to scale, raising the daily budget, breaks the platform's optimization and pushes the cost up in the moment. Spend more, pay more per user.
- Without the right setup, the average cost drifts well above the line. Single approaches could beat the target, but each new out-of-range test dragged the average back over it.
- So three things had to be solved at once: an approach that was stable, one that was cheap, and one that was good — plus the settings to run them on.
Under all of it, the missing piece wasn't a tool or a channel. It was a process. Decisions were made in the moment — "what do we launch now?" — with no backlog of tested approaches to pull from and no plan reaching more than a few days out.
What we did
Our buying lead ran the full buyer route himself, end to end, found where the buying process leaked, and worked out how to close each gap — before asking anyone else to run it.
He proposed building a backlog of proven and new approaches, costed both the tests and the budget to scale the ones that hit target, and personally delivered a stable flow of new first deposits on the product. The system started as one person's provable result, not a plan on a slide.
He assembled the buying team, built the processes and dashboards they run on, and helped them adapt. In parallel he set up synchronized work with the creative team, so new material is ready before it's needed — two departments running as one line.
The core setup is diversification: spread buying across ad accounts, campaigns, approaches, optimization settings and target-cost caps, with each account kept inside a limit where its price stays stable. Each scheme needs about 50 conversions to leave the platform's learning phase, so accounts are sized to reach that instead of starving. When an account's cost starts trending the wrong way, several fresh approaches come in from the backlog and the best one stays.
What changed
The channels were never the missing piece — Meta and Snapchat were already there. What was missing was the layer between the target and the spend: a backlog of approaches to draw from, accounts run inside limits where the price holds, and a rotation that swaps a fading approach before the average moves. Buying that reacts in the moment can hit a number once; buying that runs on a pipeline holds it at volume, month after month. That's the layer we built.
Trying to scale paid buying without pushing your cost per acquisition over the line?
Write us what hurts. Within 24 hours: an honest answer whether we can help, and the next step if we can.
Get in Touch