A few years ago, this conversation would have centered on sustainability, wellness, and purpose. Today, with cost of living pressure, AI powered discovery, and shopping journeys that increasingly start outside the retailer's own channels, the question is sharper. Do consumers even know what they want anymore, and how should retailers and brands respond?
At Retail Innovation Forum London 2026, the Innovation Focus panel explored this question across value, agentic commerce, personalisation, loyalty, and innovation discipline. The conclusion was clear: technology is transforming how people discover and buy, but the fundamentals of consumer behaviour have not changed. Trust, relevance, value, and simplicity continue to decide who wins.
Six themes stood out:
Value sets the boundary, but values still decide
"Price is always going to really seriously matter. But within those boundaries, people still care."
Consumers are more cost conscious and more cautious than they were two years ago. Down trading is real, discretionary categories such as home, garden, and general merchandise are softer, and shoppers are protecting categories that bring them joy by saving less and borrowing more.
Within those price boundaries, values still drive choice. Health, sustainability, quality, and convenience act as tie breakers when core needs are met, and shoppers increasingly expect credibility rather than slogans. The opportunity is not to charge a premium for purpose, it is to use better product data to surface the right product at the right price.
What this means: Win on price credibility first, then use trustworthy product data to differentiate on values. Treat values as a tie breaker engine, not a price lever.Agentic commerce is real, but filtering comes before full delegation
"You can reduce your candidacy of what you need to decide between. I think they're quite good for that."
Traffic to Gen AI browsers has reportedly grown more than 5,000% in the US, and consumers are clearly willing to let AI narrow their choices. They are not yet willing to hand over the entire purchase. Browser-based shopping agents are slow, incomplete, and uneven across retailers, and full automation remains a poor experience.
The near-term opportunity is to help shoppers filter and decide faster, not to replace the decision. The risk is also clear. Hallucination becomes a financial risk when an AI assistant promises returns, prices, or availability the operation cannot deliver. Retailers need deep integration of inventory, pricing, returns, and product data into any AI surface they expose.
Open standards such as Model Context Protocol make this dramatically cheaper than it used to be. As one panelist put it, "if you don't invest in that now, then it's a complete wasted opportunity." Watching and waiting may be the riskiest strategy of all.
What this means: Invest in AI ready infrastructure now. Prioritise accuracy, real time data flows, and open protocols over flashy front ends.Proof beats persuasion
"What we want is the facts."
As more discovery moves through AI, the quality of the underlying product data becomes the brand. Marketing copy designed to persuade does not build trust with an agent that is trying to compare and rank. Verified claims, structured proof, and trusted data sources do.
This shifts the competitive frontier. Brands that invest in credible, machine-readable proof become more visible to agents and more defensible to shoppers. Brands that rely on persuasion alone risk being filtered out before a human ever sees them.
What this means: Treat product data as a strategic asset. Audit claims, structure proof, and prepare for a world where AI evaluates your brand before your customer does.Personalisation must earn the data it uses
"People are happy to do it, as long as they're getting value."
The fundamental value exchange has not changed. Consumers will share data when they receive clear, immediate utility in return. What has changed is the expected standard. When an AI assistant already knows a shopper has children and still recommends an adult-only destination, the experience feels broken, not creepy.
Recency and real time context now matter more than long historical profiles. Implicit signals such as the product being viewed, the basket being built, and the question being asked predict intent better than persona data alone. The risk is no longer over personalisation, it is under personalisation that ignores the context the consumer has already provided.
What this means: Move from profile based to context-based personalisation. Use behavioural and basket level signals in the moment and be transparent about who is responding.Loyalty is a repertoire, not a relationship
"Most retail brands within your category, you're one of seven brands that the customer's buying from."
Customers do not give exclusive loyalty. Even the most loyal shopper in a category typically buys from a repertoire of about seven brands, and the composition of that seven shifts every year. Baskets are shrinking, channels are multiplying, and AI agents will accelerate this fragmentation by routing fulfillment across multiple retailers based on rules and context.
That makes loyalty a continuous earning exercise across every interaction, not a points balance. It also raises a strategic question for retailers. As one panelist framed it, "call me what you like as long as you call me." Brands and retailers will need to decide whether to invest in being chosen on their own surfaces, on someone else's surface, or both.
What this means: Compete for repeat selection, not exclusivity. Align acquisition, retention, and service under one customer view, and prepare for loyalty to shift toward the agent the shopper trusts most.Innovation discipline beats innovation volume
"Experimentation is really important."
Innovation costs have fallen dramatically. Prototypes can be built in an hour, synthetic users can stress test concepts before real consumers see them, and AI tools have replaced large parts of traditional agency work. The constraint is no longer access to tools; it is strategic focus and measurement discipline.
The panel was equally clear that betting on behavioural change is dangerous. Some ideas are simply ahead of their time, and being a fast follower is often cheaper than being first. The winning posture is to stay informed, build the playbook, and be ready to execute the moment a behaviour shift becomes real.
What this means: Cut the experimentation portfolio, sharpen measurement, and prebuild the execution plans for the bets that matter. Speed will come from preparation, not from launching more pilots.
In summary: the technology is new, the expectations are not
AI, personalisation, and agentic commerce are changing the surfaces of retail, but consumers continue to reward the same fundamentals they always have. Trust, relevance, value, and simplicity decide who wins. The organisations that pull ahead in the next phase of retail innovation will not be the ones that adopt the most technology. They will be the ones that apply technology with discipline to remove friction, deliver proof, and consistently earn the next purchase.
Thank you to all involved for contributing to a thoughtful, experience‑led discussion on what it truly takes to move from AI experimentation to enterprise‑wide impact.
Panel participants:
Darren Wright, Reward
James Taylor, Particular Audiences
Benni Lickfett, Diageo
Sarah Arana-Morton, Provenance
Paul Wilkinson, Lloyds
Moderator:
Gaby Appleton, Chief Product Officer, dunnhumby
Continue the conversation at Retail Innovation Forum Americas 2026
For those interested in exploring more insights like these and continuing the conversation on innovation readiness, ecosystem collaboration, and the future of retail, we will be hosting the Retail Innovation Forum Americas 2026 on September 10, 2026 in Bentonville, Arkansas. The Forum will bring together retailers, brands, startups, and investors to share practical perspectives on how emerging technologies are reshaping decision‑making, operations, and customer experience across the retail value chain. Learn more about the event and how to participate at Retail Innovation Forum Americas 2026.
Innovation insights powered by the dunnhumby Retail Innovation Network
The Network brings together retailers, brands, startups, and investors to turn emerging technology into practical, scalable impact. To stay connected and explore future Retail Innovation Forums, visit dunnhumby ventures.
Events like the Retail Innovation Forum showcase the strength of the dunnhumby Retail Innovation Network, the world’s fastest-growing open innovation community for retail technology. With members spanning retailers, brands, startups, and investors, the Network is driving meaningful progress across the industry. We extend our sincere thanks to all participants, speakers, and attendees who made this event a success.
To receive insightful thought leadership and engagement, explore joining the Retail Innovation Network, the fastest growing open innovation networks in retail, for free at www.dunnhumby.com/innovation and find out more about the next Retail Innovation Forum at Retail Innovation Forum Americas 2026.
