In my experience, sustainable advantage rarely comes from a single brilliant campaign. It comes from building a system where good decisions happen repeatedly — where data is trusted, teams are aligned, and technology supports the outcome rather than dictating it.
That’s when AI stops being a pilot project and starts becoming real infrastructure.
For much of my career, marketing rewarded craft above almost everything else. The best copywriter. The best campaign idea. The leader who personally reviewed every asset before it went live.
For a long time, that model worked because marketing operated at a human scale. Campaigns were periodic, channels were manageable, and decisions moved at a pace where individual oversight was possible.
That environment no longer exists. AI is now embedded across marketing operations — from targeting and media optimisation to content generation and customer analytics. A recent report by Gartner noted that roughly two-thirds of CMOs expect AI to materially reshape their role in the next few years.
Yet adoption alone is not translating into impact. McKinsey’s global research shows that while AI use is widespread, only a relatively small group of companies are capturing meaningful enterprise value. BCG’s analysis reflects a similar pattern — nearly three-quarters of organisations still struggle to scale AI impact despite significant investment.
In my experience, the limiting factor isn’t the technology. Most organisations already have the tools. The real constraint is the system those tools sit inside. When data is fragmented, identity models are inconsistent, and governance is unclear, AI doesn’t create clarity — it scales inconsistency.
One uncomfortable truth in marketing today is this: AI doesn’t fix bad data. It amplifies it.
There is a lot of enthusiasm around scaling marketing with AI — more content, more campaigns, more personalisation. But often that personalisation is built on CRM data that is incomplete, shaped by legacy sales processes, and updated inconsistently across teams.
Many organisations are celebrating productivity and output velocity while quietly watching pipeline quality decline.
Before scaling an AI-driven marketing motion, I believe leaders should ask a few simple questions:
- How old is the data your AI models are learning from?
- When was your CRM last audited for completeness and accuracy?
- Can you trace AI outputs back to validated insight — or just confident assumptions?
These questions are uncomfortable, but they matter. In my opinion, the role of marketing leadership is shifting from execution to orchestration. The job is no longer:
- Writing the best copy
- Producing the best campaign
- Personally reviewing everything
The job now is:
- Designing the system that produces quality consistently
- Ensuring data foundations and governance make decisions credible
- Deciding what must remain human — judgment, ethics, creativity
- Knowing where automation accelerates outcomes, and where it risks eroding trust
Despite all the technology, the fundamentals of good marketing remain unchanged. Listening, learning and acting on what customers actually need is still what drives real progress. I’ve been fortunate to work with organisations that genuinely show up for their customers and deliver — because in the end it’s the partnerships we build and the outcomes we create together that matter.
In my experience, sustainable advantage rarely comes from a single brilliant campaign. It comes from building a system where good decisions happen repeatedly.
That’s when AI stops being a pilot project — and starts becoming real infrastructure.