18.78× on $715,000 of Ad Spend

This is the best single account we have run: $715,000 of ad spend returning $13.4M in tracked revenue, an 18.78× return. It is one account over one period, not an average — our average across managed campaigns is 8.4×. Both numbers are self-reported from ad-platform data. This page explains what actually drove it, and what did not.

01Kuwait Army05Gulf Bank06Warba Bank07Boursa Kuwait

The headline numbers

A single managed account at THE UNKNOWN BRAND returned 18.78× on ad spend: $715,000 invested against $13.4M of tracked revenue. That is the ceiling of what we have achieved, not the middle. Across all managed campaigns between 2023 and 2026 our average is 8.4×, and the honest way to read this page is as an account that went unusually right, documented so the conditions are visible.

Why we are not naming the client

We do not publish client performance data attached to a client name without written permission, and we have not asked for it on this account. Naming it would make the number more persuasive and less honest, because the client did not agree to have their revenue published. The figures are auditable in the ad account by anyone we are actually working with, and we will walk through the platform reporting in a call under NDA.

The same rule is why this site does not carry testimonials it did not collect or logos it cannot evidence. If a number matters enough to publish, it matters enough to be checkable by the people who can check it.

What actually drove it

Three things, in order of contribution. None of them is a secret, and all three are conditions rather than tactics — which is precisely why the result is not reproducible on demand.

• Creative volume against a working offer. The account had an offer that already converted before we touched it. Our contribution was replacing fatigued creative faster than the audience tired of it — because the creative is generated rather than filmed, a losing ad was replaced the same week rather than the next quarter. Volume only compounds on top of an offer that works; it cannot rescue one that does not.
• Margin that could absorb the spend. High-margin categories tolerate aggressive scaling. A thin-margin account spending the same money would have shown a fraction of the return at the same efficiency, because the ceiling is set by the economics, not by the media buying.
• An existing audience to retarget. Prospecting is expensive; returning buyers are not. This account had a warm base before the scale-up. Accounts starting from zero do not produce these numbers in the same window, and we say so at proposal stage rather than after the first month.

What this does not mean

It does not mean 18.78× is what we quote, forecast, or aim for. It is one account, at one point, under favourable conditions we did not create. Presented as a typical outcome it would be misleading, which is why the average sits next to it everywhere it appears on this site.

It also does not mean generated creative caused the return on its own. Generation changed the rate at which we could test and replace — an operational advantage, not a magic one. On an account with a weak offer, faster creative reaches the wrong conclusion sooner.

If your category is thin-margin, your offer is unproven, or you are starting without an audience, the realistic number is materially lower and the proposal will say so before you spend anything.

How the number was measured

Return on ad spend here is tracked revenue attributed in the advertising platform divided by media spend in the same period, excluding management fees. It is platform-attributed, which means it inherits every limitation platform attribution has: view-through windows, cross-device gaps, and the well-documented tendency of each platform to claim the same conversion. We report it because it is the number the client sees in their own account, not because it is a perfect measure.

We have not had these figures independently audited. Any agency publishing performance data without saying that is either audited — in which case they will name the auditor — or hoping nobody asks.

Frequently asked questions