Reclaiming Conversation in the Age of AI
Market research is being reshaped by artificial intelligence, automation, and new conversational technologies. These changes are not simply introducing new tools—they are reshaping established practices and raising practical questions about how research is done and how its value is defined.
Based on 15 in-depth interviews with senior leaders across brands, agencies, and research technology providers, this paper finds that the future of market research will not be defined by a contest between humans and machines, or by the convergence of qualitative and quantitative methods. Instead, it will be shaped by a more fundamental shift: from methods to meaning, from projects to conversations, and from gatekeeping to strategic partnership.
Conversation is the Holy Grail
Across the interviews, one theme rose above all others: conversation is the most powerful way to understand people. As Sarah Snudden, Head of Consumer Insights at JDE US, put it, “Whether it's Qual or Quant一or anything in between一I think the Holy Grail is conversation.”
This is not a new insight. Conversation has always been central to research—how researchers understand participants, how researchers work with clients, and how brands come to understand the people they serve. What is new is the industry’s ability to support conversation at different depths and scales.
Leaders described a research landscape where conversationality nowspans from deep, human-led interviews to AI-enabled dialogue embedded within large sample size studies.
This shift is not driven by novelty, but by necessity. Decision-makers demand insight that reflects real human complexity—context, emotion, contradiction—not just clean data points. At the same time, advances in AI moderation are making it possible to hold more meaningful conversations with more people than ever before. Together, these forces are pushing the industry back toward its most reliable source of understanding: conversation itself.
While conversation is often associated with qualitative research as a data collection method, leaders emphasized that it sits at the foundation of all good research.
Conversation mirrors real-world decision-making. People don’t make decisions by choosing from pre-defined options alone; they reason through choices in conversation - with themselves and with others.
Conversation builds shared understanding and supports triangulation, allowing researchers, insight users, and stakeholders to engage with insight as it forms, making findings more credible, memorable, and actionable. As Susan Fader, Founder of FaderFocus, reflected, “Clients used to be in the room. That’s where understanding really formed.”
Ultimately, conversation aligns research with how humans think, feel, and decide.
The Role of AI Moderation: Scaling Conversation, Not Replacing Craft
AI moderation- including its ability to process large volumes of conversational data - makes it possible to have many more conversations with real people at scale. It brings new efficiencies: widening access, ensuring consistency, instilling non-judgement, and freeing researchers from repetitive tasks.
Despite reserving her preference towards human-moderated qualitative research, Margaret Tso, Senior Manager Consumer Insights at Cadillac Fairview, commented that “AI moderation offers significant advantages in maintaining neutrality, consistency, and a non-judgmental approach across all interactions. This can be particularly valuable in ensuring fairness and reducing bias in data collection.”
Sarah Snudden added on AI’s speed of conversational data processing, "for those of us who had to dig through transcripts, who doesn't love a little boost on that front. Using AI to speed the plow and get to the harvest is a really nice thing.”
Beyond efficiency, several leaders noted that AI’s non-judgmental nature can also lower barriers for participants. Without the social dynamics that exist between strangers, participants may feel more comfortable discussing sensitive or difficult topics, leading to more candid input. As Wensy Chui, Marketing Intelligence & Strategy Lead at Nestlé Health Science, said, “With AI Moderation, that barrier isn’t there. There’s no judgment or stigma, so you can explore some really challenging research topics more freely.”
This shift comes with a candid reality: qualitative moderation jobs are being disrupted. Multiple interviewees noted that some qualitative moderators have lost client work as organizations bring research in-house and rely on AI platforms. “We are seeing a dramatic shift in how Qual spend is changing for AI,” said Daniel Graff-Radford, CEO of Discuss.
However, importantly, leaders did not frame this as a binary choice between humans and machines. Instead, they described a pragmatic “middle zone” where AI is not only acceptable, but preferable for handling repetition, volume, and standardization. The responsibility of meaning-making - deciding what matters, why it matters, and how it connects to real-world decisions - remains firmly human.
The Lines Between Qualitative and Quantitative Are Blurring - or Are They Really?
While AI is making traditional distinctions between qualitative and quantitative research more fluid, few leaders believed the two were converging into a single method. Instead, important differences remain, in structure, depth, duration, and flow. New methodologies have emerged on both sides of the qual/quant divide.
Several leaders framed the new landscape as a spectrum:
Deep Quant — Conversational Quant — AI-Moderated Qual — Deep Qual
Conversation acts as the connective tissue across this spectrum. It allows researchers to choose the right balance of depth, scale, and structure for the decision at hand - without erasing the distinct strengths of each approach.
Andrea Scheuerman, VP, Insights Consulting at Prodege, reflected, “they are generally considered together, but if I think more deeply about it, AI-moderated qual is a longer, deeper conversation that unfolds organically and loosely while conversational quant tends to be embedded in a quant study, uses a consistent initial prompt followed by a series of custom probes based on the initial response. As a 'quantie' I'm more comfortable with the latter.”
Tasneem (Tas) Dalal, Customer Success Director at Product Hub | MMR Research, added, “In my experience, both tools, conversational survey and AI-moderated qual, serve distinct roles, and when used thoughtfully alongside one another, they contribute to a more complete and meaningful understanding of the consumer or user perspective.”
Conversational Quant
Industry leaders were unanimous on one point: AI moderation has clear value in extending conversational depth into places where it has historically been difficult - or impossible - to achieve, particularly within large-scale quantitative research.
For decades, quantitative surveys have delivered speed and comparability, but only within the bounds of what researchers could anticipate in advance. AI-enabled conversational approaches loosen that constraint - allowing participants to explain, reflect, and elaborate without sacrificing structure.
The result is not “qual disguised as quant,” but a deeper form of quantitative research - one that captures the why alongside the what. As Romani Patel, Director of Data Science at Microsoft, put it, “The role of AI in scaling conversations for market research is not to out-human the human but to make it possible to expand the reach of qualitative depth into traditionally quantitative spaces. Think of it as qualitative insight at scale that feels conversational but still delivers structured insights with the rigor of quantitative data.”
This shift also reframes how quantitative studies are designed, as Sarah Snudden noted, “quant is increasingly being built to flow more like a conversation.”
Crucially, leaders emphasized that strong conversational survey implementations remain largely human-led and tightly-governed. Researchers still define the objectives, design the questionnaires, set boundaries, and determine how insights are interpreted and applied. As Ethan Titelman, SVP, Creative Strategy & Insights at SmithGeiger Group, explained, conversational surveys “make quant research much deeper and pull the nuance and the why behind it out, but still are able to have the same kind of structure. And ask the same types of quantitative questions that we can ask in a traditional survey.”
That translates into easier client onboarding. Ethan Titelman continued; “It's exactly what we tell them (clients). It's still very human-led. It's very structured. We're not letting the AI go off in a lot of different tangents. It's not AI conducting research. Our team is still writing the questionnaire, like 85% of it.”
Conversational quantitative approaches also elevate participant experience from a secondary concern to a strategic asset. Poorly designed surveys produce disengaged respondents and shallow data; conversational formats do the opposite. When participants feel listened to rather than processed, they give more thoughtful, detailed answers, improving both insight quality and long-term panel health.
As Tasneem Dalal noted, “The way data is gathered absolutely influences the final output. Participant experience matters, because if people are disengaged, rushed, confused, or unmotivated by the experience, the data will reflect that.”
Daniel Graff-Radford predicted that “ten years from now, the companies that survive will be the ones with the best participant experience. The insights will simply be better.”
The difference is immediately visible. Ethan Titelman shared that “the first project I ever ran with this (conversational survey), I opened the verbatim file. That's when it hit me right away, within 10 seconds, like, Oh yeah, people are writing a paragraph and not just a basic few words. It was just immediate, seeing into the future…this is exactly what I've wanted for a while”.
At the same time, leaders cautioned that not all AI moderation delivers this benefit. Jason Jacobson, Senior Director Consumer Insights at Sekisui House, emphasized that poorly designed systems frustrate participants and flatten insight, while thoughtful implementations feel more responsive and human. As he put it, “Tech without thought is just tech. I have evaluated and worked with multiple vendors in the AI moderation space. I think the industry needs to slow down and test this more, truly think about the consumer and how we want them to feel.” Conversational surveys only succeed when participant experience is intentionally designed, not assumed.
AI-Moderated Qual
How is AI-moderated qual different from a conversational survey? Cate Riegner, Practice Leader, Digital Life & Commerce at Naxion, offered an insightful framing: “It's verbatim data, so it's qual data, and it's countable, but it's not statistical, because you're not sampling the way you would sample in quantitative research.”
Leaders described AI-moderated qual as particularly effective in contexts where speed, breadth, and consistency are valuable, and agreed that AI moderation is already delivering meaningful value, because as Jessie Xue, Customer Insights Manager at Sonova Group, put it, “AI is becoming very capable in the process, flow, probing, timing, etc.. AI applications today naturally come with experience.”
Tas shared her own experience where she embraces AI-moderated qual, “it works extremely well for early exploratory work, concept refinement, post-usage diaries, broad tension spotting, and quick, high-volume follow-up conversations. It widens the surface area and gives us a richer starting point. It has been especially helpful for diary studies, early exploratory conversations, rapid iteration, and large-scale follow-ups. It makes it possible to gather a significant amount of contextual insight quickly… I always add a human layer of review because nuance can sometimes be missed, but it is extremely effective for scaling qualitative inputs at speed.”
Daniel Berkal, SVP at The Palmerston Group, expected AI moderation to play a large role in moderation. He estimated that “roughly seventy percent of standard qual projects can live in that middle zone (handled by AI moderation) where the objective is simple comparison, preference, or rapid reasoning.” However, he emphasized that “for deeper, messier human work, you still need a person.” Daniel Berkal also highlighted the missing “spark” in AI: “The formulas are there, but the spark isn’t. The problem is that humans break structure in ways that feel alive. AI does not. At least not yet.”
AI can scale and streamline conversations, but it cannot replicate subtle tonal shifts, hesitation, contradiction, or humor - the moments where meaning emerges unexpectedly. As Susan Fader explained, “so much behavior is automatic. So much behavior they haven't really thought about. I call that contextual intelligence”.
Looking ahead, AI-moderated qual is expected to shift research toward more continuous, stream-oriented workflows rather than discrete projects. Cate Riegner described a future where “organizations constantly gather and integrate qualitative insights, creating a real-time feedback loop for strategic decision-making”.
In this model, AI handles volume and repetition, freeing researchers to focus on interpretation, meaning-making, and strategic guidance - the very aspects of research that cannot be fully automated. As Tom Woodnutt, Founder of Feeling Mutual Limited, succinctly put it, “While qual researchers add value through expert design and moderation, I believe it is in the interpretation stage where we make the biggest contribution clients.”
An Industry at an Inflection Point - Reclaiming Conversation as Strategic Advantage
Technology is lowering barriers to conversational data collection, but this shift brings a growing risk of commoditization. Leaders emphasized that methodological rigor alone is no longer enough in a world of faster, cheaper, “good enough” data - the industry must redefine its value.
Zontziry Johnson, Founder of MRXplorer, cautioned that “by gatekeeping so much of what we’ve gatekept, we’re actually keeping ourselves from staying relevant.” Leaders also warned that uncritical adoption of AI can backfire: poorly designed systems frustrate participants and flatten insight.
Synthetic respondents illustrate the tension. Some see value in AI-generated agents for rapid testing or early exploration, where both moderation and participation can be simulated with diverse, human-like personalities. Others warned that fully replacing real people risks cheapening insights and eroding the relationship between brands and consumers - diluting the very essence of meaningful research.
The question is not whether to use AI, but how.
The central risk is not that AI will replace researchers, but that researchers and clients allow themselves to become too removed from participants, data, and the moments where meaning emerges - “automated moderation imposes a trade-off; it offers speed, scale and efficiency but this comes at the cost of disintermediating the relationship between researcher and participant”, as warned by Tom Woodnutt.
The ultimate reward is not efficiency alone, but human capacity amplified. As Romani Patel said, “ideally, I want AI to play the role of an amplifier, removing friction, scaling access, and making the researcher’s job less about logistics and more about interpreting meaning and driving decisions.” When researchers are freed from the repetitive and the administrative, they can invest more in the elements that machines cannot replicate - judgment, empathy, curiosity, and strategic insight. This amplified human role becomes a core competitive advantage, distinguishing organizations that can turn data into decisions, and conversation into action.
The organizations and researchers that will thrive are those that embrace AI thoughtfully—integrating it to broaden reach, increase engagement, and scale meaningful interaction, while keeping humans central to the process. As Daniel Graff-Radford observes, “Researchers who have the flexibility to choose their starting point, rather than being locked into a prescriptive journey, gain real superpowers. They can follow curiosity, deepen understanding faster, and arrive at clearer answers to the ‘why’ behind decisions—often uncovering insights that a rigid, single-method approach would miss.”
Researchers are not just moderators; they are interpreters, storytellers, and strategic partners. The craft of turning dialogue into decisions has never been more important. AI can magnify its reach, but only human judgment can preserve its relevance and spark. The future of market research belongs to those who reclaim conversation as a strategic advantage—and use AI to extend its power.
Key Takeaways
Advances in AI and conversational technologies are reshaping the future of market research. Insights from conversations with 15 industry leaders highlight three fundamental shifts defining this new landscape:
From Data Collection to Conversation: Market research is returning to its most reliable source of understanding: conversation. As decision-makers demand insight that reflects real human complexity, AI moderation is enabling meaningful dialogue at unprecedented scale. Conversation mirrors how people actually reason and decide, moving research beyond pre-defined response options and enabling shared understanding as insight forms. In this new paradigm, conversation is both the medium and the differentiator.
From Methods to Meaning: AI moderation excels at volume, consistency, neutrality, and speed, creating better participant experience and conversations at greater scale. Qualitative and quantitative research are now connected by conversational approaches, but remain distinct: conversational survey uncovers “the why” with quantitative rigor, AI-moderated qual handles higher-volume, or repetitive exploratory tasks. Freed from routine work, researchers can focus on deeper, messier work where empathy, intuition, and contextual intelligence matter most.
From Gatekeeping to Strategic Partnership: As AI lowers barriers to conversational research, research value shifts toward interpretation, story-telling, and strategic partnership. The risk is not automation itself, but commoditization and detachment from the human moments where meaning emerges. The future belongs to teams that use AI to remove friction and expand reach while keeping humans central as interpreters, sense-makers, and decision partners. Researchers evolve from gatekeepers of methods to strategic partners who turn dialogue into decisions.
Methodology: This paper draws on 15 in-depth interviews conducted in Fall 2025 with industry leaders across the market research ecosystem, from brand-side to agencies to technology/solution providers. Interviews were mostly conducted one-on-one via online video, lasted approximately 30 minutes and did not use AI moderation. The choice of methodology reflects the study’s objective: to capture expert reflection, professional judgment, and real-time sensemaking from experienced practitioners through direct human conversation. Some of the quotes were lightly edited for length and clarity.
The interviews were conducted by Kathy Cheng, Founder & CEO, and Tim Benner, Head of Customer Success, at Nexxt Intelligence | inca.
The article was authored by Kathy Cheng. Kathy Cheng founded Nexxt Intelligence with 20+ years of global market research experience, and a conviction that quantitative research could be improved by adding a qualitative dimension to become more engaging, enabling and exploratory. This belief led to the creation of inca, Nexxt Intelligence’s conversational surveys that deliver robust KPIs, higher quality data, and deeper human insights through a qual-like experience. Kathy had worked with Nielsen in Shanghai, and with Ipsos and Environics in Canada. Kathy is a frequent presenter and speaker at ESOMAR, IIEX, Quirk’s, CRC, CRIC, etc., and served as a panelist on the CEO Forum at the 2024 ESOMAR Congress. Kathy is an Insights Association Laureate, Class of 2024.
This article was originally published on Greenbook. You can read the full version here: