product · 6 min read
The number that kills an ad: how auto-kill thresholds actually work
Last updated: June 2026
Fast answer
Most first ads lose money because nothing decides when to stop them. The fix is a pre-set kill number: a cost-per-acquisition band for the niche and a minimum click-through rate. CommonWealth Ops uses sourced niche benchmarks and pauses any ad that crosses out of range automatically, before it eats the budget. The threshold is the mechanism — not a feeling about the ad.
An ad is not killed by an opinion
The expensive moment in paid acquisition is not the launch. It is the silence afterward — the days where an ad quietly spends while nobody decides whether it is working. The beginner runs it "to see what happens." The disciplined operator decided what would happen before the ad ever went live.
That decision is a number. Not a hunch about the creative, not a feeling about the product — a threshold that, once crossed, ends the test. The whole skill of not losing money on ads compresses into setting that number correctly and then actually enforcing it.
Which number, exactly
There are two, and they are read together.
The first is cost-per-acquisition against the niche band. Every niche has a range where a dropship winner can profitably acquire a customer. In fitness, the internal corpus puts that band at roughly EUR 8 to 14. An ad whose CPA drifts above the band is losing money on every order it brings in — no volume fixes that.
The second is click-through rate against a floor. The vault default kill-CTR is 1.3%, with winning creative starting around 1.7%. CTR is the earliest signal you get: an ad below the floor is failing to earn the click, and spending more will not rescue it. It tells you to stop before the CPA data has even finished arriving.
Why the numbers are sourced, not invented
A threshold is only as trustworthy as where it comes from. These are priors drawn from real benchmarks — the fitness ROAS-breakeven floor of 1.9 is corroborated by Triple Whale's 2025 dataset (Health and Wellness at 1.50, Sports and Outdoors at 2.28, across roughly 35,000 brands). They are calibration parameters, clearly labelled, that your own tests overwrite the moment you run them. The benchmark gets you to a sensible day-one decision; your data takes over from there.
How CommonWealth Ops fits
CommonWealth Ops turns the threshold into an automatic loop. It watches each ad's cost-per-acquisition and pauses the ad on its own when it crosses out of range — no sitting in front of the dashboard. When exact pixel attribution is not available, it closes the loop on an estimated signal and labels it as estimated; an honest cut on a clear estimate protects more capital than waiting for a perfect number that may never arrive. And whatever happens, it never spends above your capital band's daily cap, so the damage of any single test is bounded by design.
It does not sell magic ads. Ads will still fail — that is what testing is. What changes is that each failure is caught at the number you set, not weeks later.
The next step
If you recognise the pattern of letting an ad run longer than you should, the kill threshold is exactly the discipline this system makes automatic. Alvaro is the first operator running this loop now, on the EUR 49/month plus 20% of net profit model (and EUR 0 in any month without profit). If you want the first real operator data when a slot opens, join the waitlist and see how it works for operators.
Frequently asked questions
- What number actually kills an ad?
- Two, read together. First, cost-per-acquisition against the niche band — in fitness the internal corpus puts the dropship-winner CPA at roughly EUR 8-14; an ad drifting above the band is losing money on every sale. Second, click-through rate against a floor — the vault default kill-CTR is 1.3% (winning starts around 1.7%). An ad below the CTR floor will not become profitable by spending more, so it is stopped.
- Why not just decide by hand?
- Because the decision fails exactly when it matters. Watching every campaign daily is exhausting, and the human instinct is to give a losing ad one more day in case it rebounds. A system applies the same number every time without fear or hope. The discipline only holds when it is not optional.
- Where do the threshold numbers come from?
- From sourced niche benchmarks, not guesses. The fitness ROAS-breakeven floor of 1.9, for example, is corroborated by Triple Whale's 2025 benchmarks (Health and Wellness 1.50, Sports and Outdoors 2.28 across ~35k brands). They are priors that your own real tests override the moment you have them — a starting calibration, never a promise.
Become an operator
Stop guessing what to sell.
CommonWealth Ops turns your market's competitor activity into ranked, data-backed intelligence — and protects your capital before you spend a euro on ads. EUR 49/mo + 20% of net profit. No free trial: skin in the game both ways.
Join the waitlist