TL;DR: Three answers tell you a growth marketer can’t separate a number the ad platform handed them from one the business earned. A cherry-picked ROAS. Attribution that lives inside a single ad account. A definition of scale that just means spending more. Each has a question that exposes it.
Why do three answers tell you more than a portfolio?
A portfolio shows you what ran. These three show you how somebody thinks about what came back, and that’s the part you’re hiring. It’s the same mistake three times over, so once you’ve heard it, you hear it everywhere.
None of it is about seniority or years in the seat. It’s about whether the person can tell a platform’s report from a business outcome. If you’re still working out what the role should own, what a growth marketer owns and a product marketer doesn’t is the argument underneath that one.
Red flag 1: the cherry-picked ROAS
A candidate who opens with a 23× ROAS is telling you about one campaign, not one business. In isolation it might be true. There’s usually a Meta campaign somewhere with a small, very warm audience showing something like 22× in the platform.
The reason it shows 22× is the thirty-odd other campaigns quietly feeding that conversion. Zoom out to the whole funnel and the number turns into a wild overestimate of what that person contributed.
So the question isn’t whether the number is real. It’s why they reached for that number in this conversation. At 0:00 I put it like this: they’ve told you they took $100 to the casino and came back with $2,200 every single time. Which is preposterous.
Red flag 2: why does one platform’s data never settle anything?
Because the platform is reporting on revenue it’s also selling you, and it gets to decide what counts. There’s nothing sitting above the platforms to settle a disagreement between them. It doesn’t matter whether you sell ecommerce or B2B, and it doesn’t matter which channel. Leaning on the provider’s own numbers is the problem.
Two things go wrong, and they compound.
| What goes wrong | What it looks like in the account |
|---|---|
| The same deal, booked twice | Google and Meta both count revenue the other also counted |
| No shared definition | Each platform reads from a different source, so nothing reconciles |
Here’s the shape of it. Somebody finds you through organic search, gets picked up by the Meta pixel while they’re browsing, and converts weeks later through a direct response ad on Instagram. Meta books that as its win.
Did Meta help the conversion? Probably. Did Meta capture that person’s attention in the first place? No. At 1:16 I make the case that a marketer worth hiring can tell you which of those two happened.
What should your source of truth be?
One system, outside the ad platforms, where revenue is real. Usually that’s Stripe, or HubSpot, or whatever you genuinely close and invoice through. You pick it once, and then you pipe it back into every platform you buy from, so each of them is reporting against the same thing.
Two things happen when you do. Duplicated revenue and duplicated deals disappear, because there’s finally something to reconcile against. And the platforms become accountable to the full conversion journey rather than the slice each can see.
The mechanics are their own subject, and I’ve covered which events to send back, and which one to bid on separately. In an interview you don’t need the detail. You need the candidate to know a source of truth exists and to say why. A good answer names the system, names what it owns, and says how the data gets back into the platforms. A bad one talks about dashboards.
Red flag 3: when scalable just means spend more
Spending more does create more scale, so a candidate who says “we scaled it by tripling the budget” isn’t wrong. They’re just saying something trivial, and it stops being true at the saturation point. In a narrow channel that point arrives while the budget is still climbing.
Think about who you can actually sell to. Selling shoes, your total audience is roughly the planet, and you won’t hit saturation for a very long time. Selling a verticalised B2B SaaS product, the audience is a great deal smaller.
Run ads on LinkedIn against that smaller audience and you’ll reach everybody qualified to buy. Spend more after that and your costs go up while inbound revenue stays flat. At 3:15 I walk through why a channel as narrow as Reddit gets there very quickly.
How does a good marketer separate output from outcome?
They make one acquisition worth more than one customer. Output is what you control: spend, distribution, the number of touchpoints. Outcome is what comes back. A marketer with only the first lever will always answer “spend more”, because it’s the only lever they have.
In software the usual second lever is referrals or affiliates. Acquire one customer, and that customer brings somebody else in. On paper one acquisition starts to look more like 1.1 customers. That’s how you pull your costs apart from your growth, and it’s where I land it at 4:41.
You don’t need the candidate to say the word “referrals”. You need them to reach for something other than the budget.
| Red flag | What it sounds like | What to ask instead |
|---|---|---|
| Cherry-picked ROAS | “I got a 23× return on that account” | What did the whole account return that quarter? |
| Single-platform attribution | “Meta reported it, so that’s the number” | What’s your source of truth, and how does it reach the platform? |
| Scale means spend | “We scaled it by tripling the budget” | Where did that channel saturate, and how did you know? |
Take all three together and they describe one person: somebody who has run campaigns without ever being accountable to the money. Better to find that out in an hour than a quarter.
If you’re hiring a growth marketer into a team that has to stand up paid from nothing, the foundation and first-channel decisions make a good interview walkthrough, and the work I’ve done on those builds shows what the finished version looks like. Or send me the shortlist.
