Skip to content 
Most of the AI-in-research conversation has focused on speed and efficiency. But the bigger shift is the kind of work AI is now making commercially viable for insight teams.
Analyses that once took months can now happen in weeks. Datasets that previously couldn’t be connected can now be explored together. Research that once ended with a single project can now continue generating value long after delivery.
For the last couple of years, the AI-in-research conversation has mostly centred on efficiency. How much faster. How much cheaper. How many more data points. They’re reasonable questions, but they risk underselling what is actually changing.
The more interesting story is in scale and ambition. AI is allowing insight teams to take on work that previously wasn’t feasible at any reasonable scale. Analyses that would once have taken months can now happen in weeks. Datasets that previously lived separately can now be connected automatically. Research that used to end with a final debrief can now keep generating value long after the project closes.
Three recent client projects, all run through Nucleus, our AI hub, show what that looks like in practice.
1. Connecting a 20,000-product portfolio to consumer needs at scale
Challenge. A global home and lifestyle retailer wanted to rationalise its 20,000-product portfolio in a customer-centric way. With 80 product areas spanning seven markets, that depth of customer-driven analysis had previously been impossible at any reasonable cost. The traditional route, surveying customers and cutting the data 80 separate ways, would have taken multiple researchers many months.
What we did. AI-supported design workflows within Nucleus drafted choice drivers across all 80 product areas in days rather than weeks, with our researchers reviewing and refining each set before they went anywhere near a respondent. Synthetic groups then stress-tested the questionnaire before fieldwork, helping the team secure stakeholder buy-in for the survey design. After fieldwork, interpretation workflows analysed the quantitative findings 80 times by product area, building a custom view of who each customer group was and what motivated them. At the impact stage, AI mapped every product in the portfolio to a segment, producing a self-serve tool the retailer can continue exploring as the portfolio evolves.
What changed. The retailer can now make portfolio decisions grounded in customer needs at the scale of the actual portfolio, not a sample. Six months ago, a project of that ambition wasn’t commercially viable. Now it is.
2. Spotting cultural signals before they reach the brief
Challenge. Red Bull, working with us through Dunnhumby, wanted to inform an early-stage qualitative trends programme on movement and fitness, looking to identify new opportunities across NPD, communications and engineering. Traditionally, this type of trends scan would involve hours of manual signal gathering, with the breadth of view naturally limited by an individual researcher’s reach and bandwidth.
What we did. AI-supported design workflows in Nucleus helped frame a broader, more commercially relevant view of “movement”, going beyond sport into culture, lifestyle and behaviour. That gave the team a stronger footing for the search parameters and lines of exploration. Discovery workflows then scanned thousands of online conversations and articles to surface emerging behaviours, brand innovations and cultural shifts, both inside the category and well outside it.
What changed. The result was both broader and faster. Lateral cultural signals, the kind a single researcher would struggle to uncover through sheer capacity alone, became part of the analysis. The breadth meant the team could spend more time on strategic thinking with the client, shaping NPD and communications opportunities rather than manually gathering signals.
3. Making primary research live on past the project
Challenge. A major high-street coffee brand had just completed a substantial quantitative study on category entry points: the moments and motivations driving coffee purchases. The data was rich. But translating those structured insights into something that could fuel brand and communications development was a separate piece of work that risked getting lost in a deck.
What we did. AI-supported workflows translated each category entry point into structured search queries, turning attitudes from the survey into something we could explore in the wild. Discovery workflows mapped each entry point against real-world cultural conversations and behaviours, showing where momentum was building. At the impact stage, the entry points became platforms for brand and innovation work, with stimulus pulled from inside and outside the category and ready to use in workshops.
What changed. The brand now has a living asset rather than a single-point-in-time report. The category entry points can be revisited as culture around coffee evolves. Primary research becomes a tool the business keeps working with, rather than a deliverable that gathers dust on a shared drive.
What this all adds up to
Across all three projects, AI changed what was possible.
The retailer analysed its portfolio at a level of scale that previously wasn’t viable. Red Bull and Dunnhumby explored a broader cultural landscape than any individual researcher could realistically cover. And the coffee brand turned a one-off research study into a living strategic asset.
That’s the shift much of the AI-in-research conversation still misses. The real opportunity lies in the kind of work AI now makes possible.
That’s what Nucleus is built for: opening up projects that previously weren’t feasible, not doing the same projects faster.
Want to talk through what AI could open up in your own insight programme? Get in touch.
Webinar on-demand
Beyond speed: How AI is broadening insight's commercial value
To learn more about Nucleus, our AI hub, watch our webinar recording where we explore:
- Where the real value of AI lies, beyond speed and cost.
- Why briefs that were impossible a year ago are now on the table.
- Case studies and real-world applications with leading brands.
- How this changes the conversations insight teams have with the board.
- What changes when teams move from “what can we afford?” to “what do we need to know?”
About the author
- Hasdeep Sethi
- Group AI Lead & Data Science Director, STRAT7
Hasdeep is a data science and AI leader with 10+ years’ experience building, delivering and setting strategic direction on data science, machine learning and AI projects. He helps lead STRAT7’s AI offer through Nucleus and the AI Innovation Lab, and speaks regularly at industry events including MRS, GRBN, IIEX and ASC. His particular interest is how AI and synthetic data are reshaping insight and research.