by What six MRMW Europe 2026 conversations reveal about moving from research delivery to influence, adaptability and anticipation.
Producing good research is no longer the whole job.
Across recent conversations with insights leaders from Haleon, AXA, eBay, Warner Bros. Discovery, Oxford Economics and De’Longhi, very different business challenges keep pointing towards the same underlying shift.
Research teams can generate, analyse and access information faster than ever. At the same time, the decisions they are expected to support are becoming more complex: consumer journeys are fragmented, markets overlap, AI is accelerating workflows and external shocks can reshape assumptions almost overnight.
The result is a broader definition of what a high-impact insights function needs to do. The value increasingly lies not only in producing evidence, but in knowing how to connect it to decisions.
Here are five shifts emerging from those conversations.
1. From producing research to creating business impact
The first shift sounds simple, but several of the speakers describe how difficult it remains in practice: research has to change something.
At eBay, Tanya Ozuns and Emilia Torres argue that strong methodology and clear findings are not enough if the work is never understood, remembered or acted upon. Their team has therefore spent more time thinking about what happens after research is completed — including how consumer understanding can become part of the environments where stakeholders already make decisions.
Their Consumer Immersion Event is one example. Rather than presenting another report, colleagues interacted directly with consumers. The lesson for the team was that insight activation requires design, not just distribution.
Laura Christina Höing describes a similar challenge at AXA Germany. Her test for relevance is deliberately uncomfortable: are insights actually changing anything, or is the team simply producing good-looking reports?
That question has pushed the function towards activation, interpretation and strategic use — with AI helping reduce time spent on production so that more attention can go to questions with direct business impact.
At Haleon, Ammar Basit brings the same issue into analytics and marketing effectiveness. Building measurement models is only part of the work. The real value comes when analytics travels through the organisation, connects to meaningful KPIs and ultimately influences growth decisions.
Across all three examples, delivery is no longer the endpoint.
2. From service provider to strategic partner
Getting closer to decisions also changes the position of the insights team inside the organisation.
Basit describes the journey at Haleon as a move away from being viewed as “data crunchers” and towards becoming strategic advisers. That credibility was not automatically granted. It had to be built by entering business conversations, taking stronger positions and demonstrating how analytics could contribute to growth.
At AXA, Höing describes the function as a strategic bridge-builder. Rather than operating within a research silo, the team connects data, customer understanding and decision-making, bringing together stakeholders who might not otherwise work closely together.
De’Longhi’s Stefania Accardo takes that idea further in a global context. For an organisation trying to remain globally connected while still locally relevant, she sees insights becoming the connective tissue between markets, headquarters and the wider data ecosystem.
These are different organisations, but the direction is similar. The insights function is moving away from a model in which the business brings a question to the research team and waits for an answer.
The emerging role is more embedded: connecting evidence, people and business priorities before and during the decision itself.
3. From static models to adaptive intelligence
The systems used to understand consumers also need to reflect a much messier reality.
At Warner Bros. Discovery, Iva Kralj-Taylor’s research into content discovery found that viewers encounter six to seven touchpoints before deciding what to watch, while 75% of those journeys are unique combinations. More than 90% involve multiple touchpoints.
That makes the idea of attributing a decision neatly to one isolated channel increasingly difficult to defend.
The implication goes beyond entertainment. When behaviour becomes fragmented, the frameworks used to explain it need to evolve too.
Accardo identifies a related problem across markets. Consumer behaviour no longer follows clean geographic boundaries. Trends, retailers, reviews and digital channels move across borders, making the traditional distinction between “global” and “local” less useful than it once was.
Oxford Economics adds another layer of uncertainty. Martina Bozadzhieva argues that when policy decisions, tariffs or geopolitical shocks can quickly affect demand, pricing or category dynamics, relying on a single forecast creates false precision. Scenario planning instead acknowledges uncertainty and prepares the organisation for several plausible futures.
In each case, the challenge is slightly different. But the lesson is similar: insights functions built around stable assumptions will struggle in environments that no longer behave predictably.
4. From AI speed to trusted judgement
AI appears throughout these conversations, but none of the speakers treats speed as the final objective.
At eBay, faster synthesis and easier access to information create an unexpected problem: when insight becomes easier to generate, it can also become easier to overlook. More information does not automatically lead to better decisions.
Höing makes the same distinction around AI fluency. Knowing how to use the technology is only the beginning. Researchers also need to question what an output is based on, where bias might exist and what may be missing from the underlying data. Speed only creates an advantage when the foundation is reliable.
For Kralj-Taylor, this increases rather than reduces the importance of analytical rigour. In a world where a chart can be produced in seconds, understanding what the evidence genuinely supports — and what it does not — becomes more valuable.
And at De’Longhi, Accardo frames the issue in terms of trust. If data was the defining asset of the previous decade, she argues, trust is becoming the critical one now: trust in the systems, but also in the people interpreting and using them.
The common message is not anti-AI. Quite the opposite. AI can free research teams from parts of the production process.
But as production gets easier, judgement becomes more visible as the differentiator.
5. From reactive research to anticipation
Perhaps the biggest shift happens before a project even begins.
Kralj-Taylor wants insights teams to spend less time simply responding to questions and more time shaping the questions the business should be asking. Her phrase “before the brief” captures the ambition: move from reactive support towards anticipation, foresight and a stronger point of view.
Bozadzhieva approaches anticipation from a different direction. Scenario planning asks teams to prepare for multiple plausible outcomes before disruption arrives, rather than waiting for a shock and then explaining what happened.
Accardo’s global-local model requires something similar: an insights system capable of spotting signals across markets and deciding when a trend, behaviour or piece of evidence is relevant somewhere else.
Even Basit’s approach to European market research reflects this thinking. Rather than automatically commissioning the same work country by country, Haleon groups markets with similar consumer and shopper behaviours, researches the most relevant examples and transfers learning where appropriate.
Anticipation therefore does not only mean predicting the future. It also means building an insights function that is better prepared for what comes next — and less dependent on waiting for the next request.
The next advantage may not be more insight
Taken together, these six conversations point to a profession expanding its remit.
Research quality remains fundamental. But the strategic advantage increasingly comes from everything around it: understanding where evidence can influence a decision, building trust with stakeholders, adapting frameworks when reality changes, challenging assumptions and helping the organisation see what it may need to understand next.
That may be especially important as AI continues to reduce the friction involved in producing information.
If generating insight becomes easier, the harder questions become more valuable:
- Which evidence matters?
- Who needs to hear it?
- When can it influence a decision?
- And what should the business do next?
Those questions run through the programme at MRMW Europe 2026, taking place on October 21–22 in Berlin, where insights leaders from Mastercard, adidas, Unilever, REWE Group and Nestlé Nespresso will join the wider research community to explore how the role of insights is changing in practice.
MRMW Europe 2026 | October 21–22 | Berlin. Explore the programme and continue the conversation in Berlin.
