24 April 2026

From optimiser to operator

In summary

• What: The paid media role is shifting from execution to system ownership.
• How: Automation is doing the optimisation, pushing teams upstream.
• Why: Performance now depends on inputs, structure, and governance.
• When: Already happening, most teams just haven’t called it yet.

When optimisation actually meant something

There was a time when being good at paid media meant knowing what to tweak. You’d go into an account and make constant changes, adjusting bids, shifting budget, refining audiences, pausing what wasn’t working, scaling what was.

And it worked, because the platforms needed you to do that. They weren’t especially smart, and performance was tied directly to how actively someone was managing the account.

That’s changed.

Most platforms are now designed to optimise themselves. Bidding strategies learn. Audiences expand. Creative rotates. Performance is modelled in real time. You can log in, make very few changes, and still see results move.

Which sounds like progress. But it raises an uncomfortable question: if the platform is doing the optimisation, what exactly are we doing?

The role hasn’t disappeared, it’s moved.

This is where a lot of the industry is out of sync with reality. We’re still talking about “optimisation” like it sits inside the platform, but most of the meaningful work has already moved somewhere else.

Upstream.

Into things like how conversion tracking is set up, what data is being passed back, how audiences are defined and refreshed, how feeds are structured, how measurement is designed.

Because that’s what the platforms are actually learning from. And if those inputs aren’t right, it doesn’t matter how good the bidding strategy is, the system will still optimise, just in the wrong direction.

Automation didn’t simplify paid media. It hid the complexity.

There’s a narrative that automation has made things easier. In practice, it’s just made things less visible.

Before, if performance dropped, you could usually trace it back to something obvious, a bid change, a budget cap, a targeting issue. Now, performance tends to drift.

You might see conversions holding steady but quality declining. CPA stable, but revenue not following. Platform reporting strong results that don’t match internal numbers.

Nothing is obviously “broken,” but something isn’t right. And more often than not, the issue sits in signal quality, tracking gaps, audience inputs, or how the platform is interpreting your data.

This is the part that’s easy to miss, because it doesn’t sit in a dashboard.

Inputs are the real optimisation layer now

Zoom out and the job hasn’t become less important, it’s just changed shape. We’re no longer optimising campaigns in the traditional sense. We’re designing the conditions they operate in.

That means thinking more about:

• what signals we’re giving platforms
• how consistent and complete those signals are
• whether conversion events actually reflect business value
• how audiences are seeded and scaled
• what the platform is able to “see” and learn from

This is exactly the dynamic we’ve flagged in our work on signal resilience and measurement: once deterministic signals drop, platforms don’t slow down. They compensate. They model. They infer. They group users into cohorts that look right.

That can work well. But it also means performance becomes much more dependent on how well the system is set up to guide that modelling.

Structure matters more than ever (even if no one’s talking about it)

One of the biggest shifts I’ve seen across accounts is how much structure now matters. Not just campaign structure, but everything around it, how things are named, how data flows between platforms, how conversion logic is defined, how consistently things are implemented across channels.

Automation amplifies whatever sits underneath.

If the structure is clean and aligned to the business, performance tends to scale well. If it’s messy, inconsistent, or built around legacy thinking, the platform still optimises, just not in a way that’s easy to control or explain.

This is where a lot of inefficiency is creeping in right now. Not because teams aren’t working hard, but because the system they’re working within wasn’t designed for how platforms operate today.

The gap between reporting and reality is widening

Another thing becoming more obvious in 2026 is the gap between what platforms report and what businesses actually see. It’s not new, but it’s more pronounced.

Between modelled conversions, limited visibility on user journeys, and more activity happening outside trackable environments (AI tools, closed platforms, and so on), it’s getting harder to take performance at face value.

We’re seeing more questions from the C-suite:

• “Why don’t these numbers match?”
• “What’s actually driving growth?”
• “Can we trust this?”

Which is fair. Without a clear measurement layer, you’re often looking at outputs without fully understanding the inputs.

This is where the role shifts again, from reporting performance to validating it.

From optimiser to operator

This is the shift. The best paid media people right now aren’t the ones making the most in-platform changes. They’re the ones who understand:

• how the system is set up
• what data is flowing through it
• where signals are strong or weak
• how platforms are likely interpreting that data
• where the blind spots are

They’re not just running campaigns. They’re operating a system.

Louder’s recommendations

• Audit your inputs, not just your outputs: Don’t rely on platform reporting alone. Pressure-test your tracking, signals, and conversion logic first, that’s what everything else is built on.
• Align conversion events to real business value: If platforms are optimising toward the wrong goal, performance will look good but won’t translate commercially. Define what actually matters.
• Simplify and standardise account structure: Reduce fragmentation, improve naming conventions, and remove legacy complexity.
• Strengthen your signal quality: Invest in first-party data, server-side tracking, and consistent event capture. Gaps here have a compounding impact on performance.
• Separate reporting from validation: Platform numbers are a view, not the source of truth. Use incrementality testing and broader measurement frameworks to understand what’s really driving outcomes.
• Treat paid media as a system, not a channel: It’s a significant investment, but when media, data, and measurement are connected properly, it drives compounding return over time.

Get in touch

Get in touch with Louder to discuss how we can assist you or your business and sign up to our newsletter to receive the latest industry updates straight in your inbox.