Your Team Culture Shows Up in Your ROAS: Why We Run Ours Like a Media-Buying System

When was the last time your agency's "culture" actually showed up in your ad account? Not in a nice deck or a LinkedIn post about team values. In the numbers. In how fast someone caught the ad that was quietly bleeding budget on a Sunday, or how quickly a tired angle got replaced before it dragged your ROAS down a week.

I ask because most founders never connect those two things. Culture lives in one box, performance in another. And I think that split is exactly why so many brands end up paying a big agency with a great vibe and getting mediocre results.

Here's my take, and it's not the comfortable one. Your team's culture is a performance input. It's not the foosball table or the offsite. It's the set of behaviours your people repeat every single day, and those behaviours land directly on your ROAS, your cost per acquisition, your retention. So we stopped treating culture as an HR thing and started running ours like a media-buying system: measured, with clear standards, watched as closely as we watch an ad account.

Let me explain what that actually means, because I think it changes how you should pick who runs your spend.

Why happy hours don't move your ROAS

There's a whole genre of agency culture that's basically theatre. Team lunches, motivational speakers, a Slack channel full of celebration emojis. None of it bad. Almost none of it load-bearing.

The problem is that an inspiring talk or a fun event is an external thing, dropped in from outside, that fades by the time everyone's back at their desk. Years of habitual behaviour will always beat a one-off moment of motivation. If your team's default behaviour is slow and reactive, no offsite fixes that. You'll just have a slow, reactive team that had a nice day out.

I believe culture only counts when it's embodied in what people actually do with each other and with your account. So the question I care about isn't "is the team happy?" It's "what does the team reliably do, every week, without being told?" That's the thing that compounds into results, and it's the thing you can actually measure.

And that reframe matters more than it sounds. Once you treat culture as measured behaviour rather than mood, you can design it on purpose. You can put it on the balance sheet, in the sense that a team built this way is genuinely worth more to a brand than a bigger team that isn't. It's an asset. It just doesn't show up with a dollar figure next to it.

The behaviours we actually measure

So what do we watch instead of vibes? A handful of specific, boring things. Here's my take on each:

Response time. When Meta does something weird overnight, or a brand sends through a fire at 4pm, the gap between "it happened" and "someone's on it" is a performance number. A slow account is an expensive account. We treat speed of response as a standard, not a personality trait, because it directly protects spend.

Testing cadence. How many genuinely different angles hit the account this week, every week. Not a flurry when results dip and silence when they're fine. A steady, relentless drumbeat of testing is what keeps an account alive, because the platform is an AI that needs constant fresh inputs to keep finding your buyer. If the cadence slips, you can watch performance drift a fortnight later. So we measure it like we'd measure anything else that matters.

Creative-review rituals. Before anything goes live, it gets a proper review against a clear standard, not a shrug and an upload. Does it carry a real angle. Does it handle an objection. Is it different enough from the last six we shipped. A review ritual catches the lazy ad before it spends your money, not after.

None of these are exciting. That's the point. Excitement isn't a strategy. Repeated, measured behaviour is. And when those behaviours are the standard rather than the exception, the account just runs better, week in and week out.

A small senior team beats a big junior one

This is where I'll plant a flag. For performance marketing, a small team of senior people who've actually run accounts at scale will, in my experience, beat a big agency with layers of juniors and a nicer office. Most of the time, comfortably.

The reason is structure, not heroics. On a small senior team, the person making decisions on your account is the person who knows what a 1.8 ROAS climbing to 3.2 actually requires, because they've done it before. There's no game of telephone between a junior who touches the account and a strategist who never logs in. Fewer people, more skin in the game, faster decisions.

A media buyer I respect, who's spent over a billion dollars across his career, runs his whole shop with about five people on purpose. He's turned down chances to become a big agency again and again, because he knows that past a certain size you lose the thing that made the work good. I think he's right. The win isn't scale of headcount. It's keeping senior judgement close to the spend.

There's a second trick in there worth stealing. He deliberately has different people look at the same accounts from different angles, one in the weeds day-to-day, another higher up with fresh perspective, then they swap. Because if one person lives in an account 24/7 with nobody else looking, they lose perspective. Building that cross-look into how the team operates is a design choice, and it's a cultural one. It only happens if your team trusts each other enough to have their work examined without it turning into a fight.

What this looks like when it works

Let me give you an anonymised picture of how team design shows up in a result, because the thesis is abstract until you see it land.

Take a homewares brand we'll leave unnamed, sitting around $24k/month in spend with a ROAS that had gone flat and a previous agency that was, frankly, slow. Nice people, big team, glacial. The account hadn't seen a genuinely new angle in weeks.

What changed wasn't a secret tactic. It was the operating rhythm. A senior pair on the account, a testing cadence that put several fresh angles in every week, a review ritual that killed the weak creative before launch, and response times measured in hours not days. Over a few months that brand scaled to roughly $190k/month while holding its margins, because nothing was leaking and the testing kept finding new pockets of audience.

I'm not presenting that as a guaranteed outcome, and your mileage will differ. The point is the cause. The scale-up didn't come from a clever hack. It came from how the team was designed to behave, repeated consistently, while the budget grew. That's culture showing up in ROAS, in the most literal way I can describe.

Culture as a balance-sheet asset

So here's the thing I want to leave you with, because I think it reframes how you should choose an agency.

When you're deciding who runs your spend, you're trained to look at case studies and channel expertise. Fair enough. But underneath those results is an operating culture, a set of behaviours that either protects your money or quietly wastes it. A team that's senior, measured and fast is worth more than its headcount suggests. A team that's big, junior and event-driven is worth less than its pitch deck suggests. Neither shows up cleanly on an invoice.

That's why we run ours the way we do. Not because measured behaviour is fashionable, but because it's the part of an agency that actually touches your numbers. The happy team is a nice side effect. The standards are the asset.

So I'll hand the question back to you. If you looked past the friendly calls and the shared docs, and you watched only what your current team reliably does every week, would those behaviours explain your results? And if the answer makes you a little uncomfortable, what would it take to redesign the behaviours rather than just hope for a better month?