It was immediately clear that something wasn’t right.
When we recently won a new client from a large network agency and took over their paid search accounts, the problems were visible from the outset; in the account structure, in the reporting, and in the client’s lack of confidence in the work.
To be clear, this wasn’t about incompetence. Large network agencies employ talented people and deliver strong work for many clients. But we do often see recurring structural patterns when performance activity is managed at scale within the big agency groups – patterns and behaviours that can prioritise optics, agency revenue retention, and the service & delivery of major/flagship accounts over pure effectiveness.
When brand and demand are treated as the same thing
One of the most obvious issues for paid search was the lack of separation between brand and generic activity.
Jargon buster
Brand activity: Ads that target people already searching for your brand name (they already know you).
Generic activity: Ads that target people searching for the category or product, not your brand specifically (they’re still shopping around).
Brand keywords, capturing demand that already existed, were bundled into the same campaigns as non-brand, generic search terms. This made overall performance look healthier than it really was, while removing any meaningful way to understand or optimise true demand generation.
This isn’t a subtle technical oversight. It’s a fundamental error. Without a clear distinction between brand and generic activity you can’t see what’s actually driving growth and you can’t make informed decisions about budget allocation.
Strong performance and weak performance were effectively pooled together. The result was a set of numbers that looked reassuring, but couldn’t be interrogated or improved. It brought to mind the late-2000s subprime mortgage market, where risky assets were bundled with stronger ones to create something that appeared stable, but collapsed under scrutiny. When everything is blended, nothing can be properly managed.
A structure that prevented optimisation
The brand issue was symptomatic of a wider problem: the account wasn’t built to learn.
Campaign structures were blunt and inflexible. Ad copy was dated. There was little evidence of systematic testing or iteration, or of a clear learning narrative over time. It was difficult to see what had been tried, what had worked, and what had informed subsequent decisions.
This kind of setup doesn’t just limit performance, it actively prevents improvement.
In most cases, the issue isn’t capability – it’s the incentives and operating model behind the work. Large agencies are optimised for consistency, scale and efficiency. Those same strengths can work against accounts that require careful, hands-on management.
Following the money, not the effectiveness
Another red flag was the role display appeared to play in the media mix.
A significant proportion of budget had been directed into display activity, with limited evidence that this was being driven by effectiveness or clear business outcomes. While display can play an important role in the right circumstances, its prominence here raised questions, particularly given how opaque the reporting was around its contribution.
There’s an uncomfortable truth in paid media: some channels are more commercially attractive for agencies than they are effective for clients.
We take a deliberately different stance. We’re channel neutral by design. Our media plans start with effectiveness, not margin. If a channel can’t clearly justify its role in driving real-world outcomes, it doesn’t earn budget.
The real cost: loss of trust
Perhaps the most damaging outcome of all this wasn’t inefficiency, it was erosion of trust.
The client didn’t believe the numbers they were being shown. They couldn’t clearly see what was working, what wasn’t, or why decisions were being made. Reporting existed, but it didn’t support confident decision-making.
Once trust in data is lost, everything else becomes harder. Budgets are questioned. Recommendations are doubted. Momentum disappears. Paid media stops being a growth engine and becomes a source of anxiety.
Fixing the mess
Putting this right wasn’t about clever tactics or platform tricks. It was about fundamentals.
That meant properly separating brand and generic activity, rebuilding campaign structures so optimisation was possible, refreshing ad copy based on real intent, and reconstructing reporting so it showed not just what happened, but why.
None of this is fast. None of it looks impressive in a pitch deck. But without it, performance marketing simply can’t be effective.
Built for effectiveness
The lesson here isn’t that network agencies can’t do performance well. It’s that the model often makes it harder.
Performance requires attention: clear structures, disciplined testing, honest reporting and channel decisions made for outcomes, not convenience. When those fundamentals are treated seriously, effectiveness follows.
Working with Flight Feather
If you’re frustrated with a large network agency, it’s probably not because they’re incompetent.
It’s because the model isn’t built for effective and accountable performance.
If you want a partner that is, let’s talk.
