Question 1

I've been burned by an agency before. How do I judge whether numbers like these are real?

Accepted Answer

Ask to see the losers. Anyone can screenshot a winning campaign; the useful question is what got cut, when, and why. That's why the campaigns-we-killed section exists further down this page, including one that burned a large budget at an indefensible cost per conversion before it was stopped. If someone can only show you winners, you're seeing half the account.

Question 2

Are these numbers real, and where do they come from?

Accepted Answer

Most are pulled straight from Meta and Google Ads reporting at account level, with the sync date printed under the stats. Two older engagements come from my own records because those ad accounts have since closed, and the page says so rather than blending the two together. Where a number can't be pulled live it should be labelled. That's a fair standard to hold anyone's case studies to, including mine.

Question 3

Why aren't the clients named?

Accepted Answer

Because I haven't asked each of them for permission to be a public case study, and a client's ad performance isn't mine to publish. So you get the industry, the city, the problem, the fix and the unedited numbers. Names stay private until someone signs off on a full write-up.

Question 4

What counts as a good ROAS?

Accepted Answer

There's no universal number, and anyone quoting one is guessing. A 3× return can be healthy profit for a business with fat margins and a loss for one running thin. Work out your break-even ROAS first, roughly one divided by your gross margin, then judge campaigns against that, rather than against someone else's screenshot.

Question 5

What does "cost per lead" actually mean?

Accepted Answer

Ad spend divided by the number of people who gave you their details or messaged you. A real enquiry, not a click and not an impression. It's the right number to judge a lead campaign on, because it connects spend to something you can call back. Watch it alongside lead quality though: cheap leads that never answer the phone aren't cheap.

Question 6

Do these results only apply to clinics and restaurants?

Accepted Answer

No. The filters cover education, resorts and farmstays, restaurants, D2C, entertainment venues, clinics and retail. Clinics and restaurants have the deepest track record here simply because more of them came through. The underlying method transfers: right objective, an offer worth responding to, and cutting what doesn't work. It transfers even when the creative doesn't.

Real campaigns. Real numbers. No screenshots from gurus.

The campaigns we killed

Questions I get asked a lot