Case study · Cobra Secret

Same sales.
34.5% less budget.

A paid lead-gen campaign running a monthly media budget in the tens of thousands of shekels. Before our creatives, the account spent heavily and returned sales at a high cost. A new creative set went live — and the month after, it delivered more sales on a far smaller budget.

About the numbers

The client's performance data — budgets, costs and volumes — is confidential. Every number here is therefore shown as an index against the starting point, never as a real value. The ratios are exact and come from the client's PPC report; only the absolute levels have been withheld.

For example: a brand going from 10 sales a month to 23 appears here as 100 → 230. The growth is preserved precisely; the original volume is not disclosed.
Before and after our creative went liveIndex: the month before launch = 100
Metric Before After Change
CPACost per sale 100 61.0 −39.0%
SpendMedia budget 100 65.5 −34.5%
Sales 100 107.3 +7.3%
+44% in CTR — a metric measured before the click, so no landing page or form can account for it.

The same sales on two thirds of the budget — and no other brand in the account moved like that over those days.

01
What happened

An improvement that held.

10 UGC-style hooks in total, alongside AI-generated product cuts. Some of the versions that went live:

UGC · by hooksame ad body, different opening
Hook 2 · short cut
Hook 2 · short cut
Hook 3 · the thief
Hook 3 · the thief
Hook 4 · broken window
Hook 4 · broken window
Hook 8 · expensive car
Hook 8 · expensive car
Hook 9 · without burning the day
Hook 9 · without burning the day
AI · the productAI-generated, square format
The product
The product
Land Cruiser
Land Cruiser
why a Land Cruiser

Sales showed a large share of installs going into a single model — the Land Cruiser. The cut is generated end to end, so a version showing that exact vehicle and speaking to its owners was a change of prompt, not a new production — a signal from sales can be chased the same week.

How we measured — and why it isn't chance

The comparison: 30 days before the creative went live against the 30 days after, within the same ad account. The client's other brands in that account served as a control — same seasonality, same period, no new creative. They are not detailed here, for confidentiality.

CTR rose 44%. That is measured before the click, so a landing page, a form or an offer cannot account for it — it belongs to the creative.

Sales are a more restrained picture than leads: sales rose 7.3% while leads jumped 60% and CPL fell 57% — meaning lead-to-sale (L2S) dropped. What makes it a win is the budget: the same sales for a third less money, and so a CPA 39.0% lower.

And at daily resolution — leads around launch day:

Daily leads index (not sales)
mean of the 14 days before launch = 100

The sharp lift ran for about two weeks, then the creative fatigued and the daily rate returned to baseline. That is not a failure — it is how creative behaves. Every asset wears out; the only question is whether there is another ready to go up behind it. That is exactly why we plan a shoot day that yields an asset bank rather than a single asset — and the monthly figure above shows what that looks like.

Want to see what we'd test on yours?

Send over the ads running now and we'll come back with the three things we'd test first — whether you work with us or not.

Get a Free Creative Audit