Agentic commerce is reshaping how consumers choose. The UK’s opportunity to lead is now.
Opinion
The question for UK commerce media networks isn’t whether agentic AI is happening; it’s whether we’re building fast enough to be on the right side of it.
Agentic AI is changing the rules of commerce media. The way products get discovered, evaluated and bought is being restructured by AI systems that research, recommend and in some cases complete purchases on behalf of consumers.
The businesses that understand this and are building for it now will shape the next decade of commerce. Those that don’t will find themselves increasingly invisible, cut out of decision-making processes.
While the US has already scaled agentic AI across multiple business functions, integrated it into core commerce operations, and is actively moving toward system-level orchestration and monetisation, the UK is still building the foundations.
Almost half of US commerce media organisations describe AI as integrated and scaled across multiple functions. In the UK, it’s one in five.
Some 70% of US professionals are very familiar with and already using AI agents. In the UK, it’s 48%. These aren’t marginal differences. They reflect a structural gap in pace, ambition and readiness that will determine who shapes the next era of commerce media.
Agentic AI is already reshaping how commerce decisions get made. Not in a future-state, thought-leadership way. Right now, today. The question for the UK isn’t whether this is happening; it’s whether we’re building fast enough to be on the right side of it.
Where UK consumers actually are
The consumer picture mirrors today’s commercial environment. UK consumers sit in the middle of the pack, with 68% claiming they are very or somewhat comfortable with AI helping them choose what to buy or book. That’s behind the US at 75%, but ahead of Germany at 59%. On the surface, that looks like a reasonably receptive audience.
Across all three markets, fewer than one in seven consumers is comfortable with AI choosing and acting on their behalf. The dominant preference, and this holds firmly for UK shoppers, is for AI to suggest while they decide. Consumers are describing a co-pilot, not an autopilot, and 72% of respondents across the study agreed with that framing.
Retail, groceries, travel, and technology are the categories where UK consumers are most relaxed about AI assistance. They are less comfortable using it for banking, investments, and healthcare, where the stakes feel irreversible and personal. This isn’t unique to the UK, but the UK consumer’s relative caution compared to the US is a commercial constraint. Trust here is procedural: it comes from guardrails, reversibility and transparency. It has to be engineered, not assumed.
On monetisation, the data is nuanced in a way that should interest anyone building or selling in this space. UK consumers, alongside their US counterparts, are the most insistent on disclosure. Around three-quarters say it matters to know if a brand has paid to appear higher in a recommendation. Yet more than half say paid influence is acceptable, or acceptable if clearly labelled.
The implication for the UK market isn’t that monetisation is off the table. Rather, hidden monetisation is the specific trust-breaker, and the bar for disclosure and relevancy is high. Build for transparency and the commercial opportunity is real. Obscure it or make it irrelevant and the damage to trust will be swift and hard to recover.
UK investment
Investment intentions are real. The most common projected allocation, cited by 33% of advertisers, for the next twelve months falls in the £50,000 to £249,000 range, with nearly a third planning to commit £250,000 to £999,000. A further one in five is planning investment at the £1m to £5m level.
These are not test-and-learn budgets. They require internal governance, board justification and expected return, which means agentic commerce in the UK has already crossed the credibility threshold inside many organisations. The challenge now is less about unlocking the first pound and more about having the measurement infrastructure to justify the next ten.
The most popular funding source for agentic AI in commerce media, at 66% among UK advertisers, is performance and paid search budgets. That suggests the UK industry is approaching this with a performance lens, which is rational. But it also means the upper-funnel thinking across brand and awareness investment in agentic systems is being deprioritised relative to the US.
Next steps
For UK commerce media networks, there are several choices to make that will determine which side of that divide they land on; my top three include:
1: Integrate rather than fragment, building connected systems across data, decisioning and execution rather than becoming a point solution agents route around.
2: Prove rather than promise, investing in measurement and attribution now rather than letting it remain a permanent excuse for not scaling.
3: Choose transparency over opacity in monetisation, aligning commercial models with what consumers have already said they’ll accept.
Taken together, these aren’t incremental improvements. They are a different operating model built around access, influence and outcomes rather than placements, impressions and clicks.
The UK has the instincts to get this right: the demand for governance, the insistence on transparency and the performance discipline, but instincts aren’t infrastructure.
The winners in this new landscape won’t be those who wait for the market to stabilise. They’ll recognise that agentic commerce is already operational and move early enough to shape what consumers see, not hope to be found.
Paul Dahill is MD EMEA Sales at Koddi. Read the full report, The State of Agentic Commerce (media).