Skip to content 
A lot of the manual processing work involved in research can now be automated: drafting questionnaires, analysing transcripts, building visualisations, spotting patterns in data. That gives consultants more time to focus on the parts AI can’t replicate: judgement, context, commercial understanding and client relationships.
The role is shifting from execution to orchestration. More time is spent helping the business understand what matters, what’s changing and what to do next.
Will AI replace people? No. But it will change what they do.
AI changes where insight teams spend their time. As processing work gets automated, the role shifts towards interpretation, judgement and strategic direction. Done properly, that makes insight teams more valuable, not less.
How the shift actually works
Research workflows involve massive amounts of processing: writing guides, cleaning data, coding open-ends, building charts. These tasks need skill and attention, but they’re largely execution work.
Most of the lift sits in the same places: knowledge management, transcript and data synthesis, the kind of desk research that used to take a week. AI can get a research team 60 to 70 per cent of the way there on those tasks. The final stretch, the judgement and the strategic call, is still human.
AI can handle this processing. Not perfectly, and not without oversight, but fast enough and accurately enough to free up significant time.
What matters is what teams do with that time. The role becomes less about processing and more about orchestration: directing, interpreting and applying insight. AI can help generate outputs, but people still need to challenge them, interpret them and connect them back to the reality of the business.
That includes the things AI doesn’t fully understand: the context behind the brief, commercial priorities, internal pressures and the nuances that were never written down in the first place.
This is where human judgement matters most. AI processes information at scale, but it doesn’t know what matters in a particular situation, or why. Someone with experience does. They know what was tried before. They can tell when data is technically right but strategically misleading.
That’s really what we mean by “human intelligence, AI augmented”. AI helps with scale and speed. People provide judgement, context and direction.
Three families of agents
At STRAT7, this work runs through Nucleus, our AI hub. We use it to remove a lot of the manual processing work from research, so our teams, and our clients’ insight teams, can spend more time focused on interpretation, decision-making and impact.
Nucleus is organised into three families of agents, each supporting a different part of the insight workflow.
Design agents. These sit at the front of the project. They help draft proposals, run desk research in minutes, and build a first cut of a questionnaire or discussion guide. Not the finished article, but enough for the team and the client to start reacting to something real rather than a blank page.
Discovery agents. These sit closer to the research itself, helping us engage with real people at scale, while also supporting the processing and analysis work behind the scenes. The most visible is Maya, our AI moderator. She works through WhatsApp, in more than 30 languages, and can adapt her follow-up questions based on what respondents actually say rather than sticking rigidly to a script.
Other Discovery agents work behind the scenes, processing and themeing qual transcripts before researchers interrogate the findings, or screening survey panels for fraud as part of our data integrity work. Mostly, they take care of the kind of repetitive processing work that used to consume huge amounts of researcher time.
Impact agents. These change what happens once a project is delivered. Clients can interact with segment chatbots and digital twins almost like they’re speaking to someone who knows the customer base inside out. Insights can be turned into infographics for teams who prefer visual summaries, or AI-generated podcasts that senior stakeholders can listen to on the move.
The result is that insight becomes more accessible and more usable. Instead of a final report sitting untouched on a shared drive, the output becomes something the business can keep engaging with over time.
What changes for clients
The most obvious difference is speed. Not because corners get cut but because far less time gets lost to processing and production work. Tasks that used to take days can now happen in hours. That creates more space for thinking, interpretation and challenge: the parts of the work that actually help businesses make better decisions.
It also changes the scope of the work itself. AI can pull in additional sources, scan for patterns and challenge assumptions in ways that would previously have taken too much time to explore properly. You start considering angles that might otherwise have stayed in the background. Well, actually, they were probably already sitting somewhere in someone’s head. The difference is there’s now more time and capacity to pursue them properly.
The output becomes more useful too. When teams spend less time formatting slides and more time thinking about application, insights connect more directly to decisions. The work becomes less about delivering findings and more about helping the business act on them.
Three shifts in what insight is for
If we step back from the client experience for a moment, there’s a bigger shift happening here. It changes what insight work is for, and where it sits within a business.
From speed to influence. Faster timelines are useful. But not for the reason most people assume. The actual win is letting researchers into the room earlier, before the brief is written and the decision taken. Who has a voice in the call changes when that happens. So does how much weight insight ends up carrying.
From reactive to proactive. A lot of research still gets used to validate decisions that are already moving. But when AI makes it easier to scan more information, more quickly, insight teams have more opportunity to spot risks, changes and opportunities earlier. Most clients tell us that’s the role they actually want insight to play. They just haven’t had the time or capacity to work that way consistently.
From projects to continuous intelligence. Research has traditionally been project-based. A project gets commissioned, a report gets delivered and then everything pauses until the next brief arrives. What’s changing is that projects can now feed into a more continuous understanding of customers and markets as they evolve. Past research or data stops sitting untouched on a shared drive and starts compounding over time.
None of this happens automatically. AI can handle the processing and scale work. People still need to provide interpretation, challenge and direction. With that mix in place, insight becomes more like a working tool the business can keep coming back to.
Building AI-capable teams
None of this works without skill development. At STRAT7, we’re training the team to apply AI thoughtfully, critically and commercially: from prompting and evaluating outputs through to knowing where human judgement still matters most. The organisations that win at this transition will be the ones with people who know how to use AI well, not the ones with the flashiest tools.
What actually matters
The honest question to ask once AI is genuinely freeing up time is: what does the team do with it? Reinvesting it in more of the same work, slightly faster, misses the opportunity. The bigger prize sits upstream, in the foresight, scenario planning and business partnering AI cannot do alone.
Ultimately, the only test that matters is whether the work gets better.
Are businesses making better decisions? Are insight teams having more influence? Are organisations acting on research with more confidence and clarity?
AI on its own doesn’t create those outcomes. The value comes from combining faster processing and broader analysis with human judgement and commercial understanding.
That’s what Nucleus is built for. Using AI where it genuinely improves the work, while keeping people at the centre of interpretation, decision-making and client relationships.
In our next piece, we sit down with Jonathan Clough, COO of STRAT7, to hear his perspective on Nucleus and on building AI capabilities that insight teams can actually trust.
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
- Jonathan Clough
- Group Chief Operating Officer, STRAT7
Jonathan is COO of STRAT7 and led the creation of STRAT7 Nucleus, the group’s AI hub that helps clients move faster and take on work that wasn’t previously possible, with human oversight built in. He leads operational strategy across STRAT7 and the group’s six specialist agencies, and speaks regularly on AI in market research, including on the STRAT7 podcast.