There’s a number I get asked about a lot: €1 million per month in managed ad spend. People hear it and assume it means something about scale, or success, or some secret strategy.

What it actually taught me is much less glamorous: attention is the bottleneck, not budget.

The first thing that breaks at scale

When you’re managing a few thousand euros a month, you can babysit campaigns. You check in daily, you catch anomalies, you adjust bids manually, you review search terms every week.

At €1M/month, that’s impossible. The number of campaigns, product groups, audiences, and ad variations is too large for any one person to hold in their head. If your system requires constant attention to function, it will break. Not because you got lazy — because the math doesn’t work.

The first thing I did when Flowboost started managing accounts at that scale was audit every manual process. If a task required human input more than once a week, it needed to be automated or eliminated.

What automation actually replaces

People worry that automation replaces judgment. It doesn’t. It replaces repetition.

Reviewing search terms, refreshing product labels, pausing low-performers, adjusting seasonal budgets — none of these tasks require creativity. They require consistency. A script can do them at midnight while you sleep and do them better than you’d do them at 4pm on a Friday.

What automation can’t replace: deciding what good looks like. Setting the thresholds. Knowing when a number that looks bad is actually fine because of context Google can’t see.

That judgment is still yours. Automation just gives you the space to use it.

The 70/20/10 rule I landed on

After a lot of trial and error, I landed on a rough allocation for how I think about account management time:

  • 70% — structural decisions. Campaign architecture, label systems, budget allocation across products.
  • 20% — creative and copy. Ad variants, landing page alignment, offer positioning.
  • 10% — monitoring. Not babysitting — monitoring. Checking that the system is running as designed.

Most accounts I’ve seen have this inverted. Marketers spend 70% monitoring (refreshing dashboards), 20% tweaking bids manually, and 10% on structure. Then they wonder why performance plateaus.

The metric that matters more than ROAS

ROAS gets all the attention. It’s easy to measure and easy to report. But at scale, the metric I care about most is contribution margin per order — revenue minus cost of goods minus ad spend, per transaction.

A 600% ROAS on a low-margin product might be worse than a 300% ROAS on a high-margin one. ROAS doesn’t see your margins. You have to build that layer yourself.

When I onboard a new client, the first question I ask is: what’s your margin by product category? If they don’t know, that’s the first thing we fix. Everything else is downstream of that number.

What the number actually means

€1M/month sounds impressive. What it actually means is that the systems have to work, because there’s no room for them not to.

That pressure is a good teacher.