Where next for market research in an AI world?

Part three

Here’s the final part of my series, looking at the impact of generative AI on the market research industry. Let’s explore the remaining avenues for human researchers in a sector that AI will increasingly automate and strategies to ensure we stay relevant.

Market analysis template

Save time, highlight crucial insights, and drive strategic decision-making

Use template

When I raise concerns to some of my market research peers about the impact of AI on the industry, those less concerned suggest that while impressive, generative AI content tends to be generic, a little obvious, lacking nuance. I agree. 

However, my concern with AI isn’t so much with what it can do now but what it will be able to do in six, 12, or 18 months. ) It can’t yet comprehensively analyze 12 hours of qualitative transcripts. However, tools like Dovetail offer AI-driven auto transcription tools and shared transcript tagging and analysis platforms, speeding up these processes considerably.

So, I predicate my suggestions for those in the industry on the assumption that we are at the foothills of AI development, and any strategies for staying relevant must bear this in mind.

Given how central AI will be in the future, it makes infinite sense for market researchers to develop AI and data science skills to ensure they remain relevant. A better-than-average understanding of how AI works, knowing when to use AI-powered tools and technologies, and understanding their limitations positions you as an AI expert rather than the thing that AI replaces. 

The challenge here may be that user-friendly market research and insight, delivered by AI, becomes accessible to anyone with a question about their consumer or wanting to optimize their marketing approach.

Even in these circumstances, the expertise of a researcher, including the capacity to perceive, evaluate a research framework, scrutinize a suggested consumer sample (even if it is synthetic), and question the responses generated by AI, will still be enhanced by the involvement of an individual well-versed in both AI and market research best-practices.

Market researchers still have a role in generating hypotheses and translating business queries into research questions, even with the widespread use of generative AI. 

While AI can process vast amounts of data and generate insights, in its current form, it lacks an intuitive understanding of human behavior, psychology, culture, and context. 

Without clever prompting, AI won’t theorize in non-obvious or subtle ways that tend to arrive at more novel, valuable research insights. Human researchers possess the creativity and adaptability to explore unconventional angles and emerging trends. 

Researchers with expertise in market research, a grounding in current culture, and an understanding of human psychology will continue outperforming AI in the quality of the questions they ask. 

This advantage is evident both in the data collection stage and back end, with humans interpreting the results and data within a broader context than AI can replicate in its current form.

At least In the short term, human researchers provide an essential human touch, combining data-driven insights with their deep understanding of people, generating and testing hypotheses that support strategic decision-making. Still, AI is undoubtedly moving in this direction.

While AI will continue to be responsible for ever-increasing volumes of data analysis and modeling of response, there will continue to be a role for strategic thinking and decision-making, leveraging the insights generated by market research. 

Ideally, humans with a grounding in market research will continue to be needed to interpret AI-generated insights, align them with business goals, and support their implementation within relevant business units.  Again, there will undoubtedly be growth in DIY research. 

However, people with a deeper understanding of marketing, insight, and culture will still be required to transform market data and insights into actionable strategies to inform decisions that drive business success.

Even with the rise of generative and predictive AI, human research experts will remain indispensable in presenting and communicating market research findings. If I’ve learned anything from the pandemic and the rise in digitally delivered presentations over Zoom and Teams, it’s how much I’ve missed human, face-to-face contact. The prospect of being debriefed by a virtual person or avatar, devoid of human input, and over video conferencing is profoundly boredom-inducing! 

Only an experienced researcher can convey complex psychological and culturally-driven insights. Similarly, adapting one’s communication styles to diverse audiences and addressing nuanced questions and concerns will still require some kind of human input. 

This combination of communication skills and human connection means that research findings are not just generated but also understood and acted upon in a meaningful way. 

Undoubtedly, over the longer term, there will be a rise in virtual consultants and presenters using automation and AI to deliver research findings; however, this feels a bit further away compared to other areas. If and when it does occur, it begs whether it will be a virtual avatar presenting to a room of equally virtual,  AI-based employees. (Will the last human left in the office please switch off the lights!)

Client-side researchers with seniority may benefit more from AI-powered market research than more junior or agency-based researchers. For instance, internal researchers will have a deeper understanding of their organization's strategic goals, enabling them to harness AI's data-driven insights for informed decision-making and business-relevant questions.

Conversely, AI is already taking a more significant role in much of the ‘bread and butter’ junior market research roles, e.g., data gathering, qualitative coding, basic reporting, and analysis. 

While arguably mundane, these tasks provide a crucial grounding in market research as a discipline. 

It’s a considerable risk to the industry and the development of juniors into future senior researchers if they don’t understand what’s ‘going on under the hood’ of the research process.

2023 survey findings from the New Market Research suggest that focus group moderation will be the least impacted research type. 

Research amongst market research and insight professionals conducted by the New MR highlights that activities like focus group moderation will be the least likely to be impacted by AI relative to other activities. 

In the previous article, I highlighted the growth in using chatbots for more straightforward, qualitative data collection. However, specific research tasks such as moderating a focus group require complex human skills that AI will continue to struggle to replicate. 

A human moderator can understand group dynamics, empathize with participants, adapt to their emotions, and foster open discussions. Human moderators can also interpret non-verbal cues, manage conflicts, and guide conversations effectively. 

AI, however, will continue to lack the emotional intelligence, empathy, and ability to comprehend subtle nuances in human interaction present in a group discussion. 

For instance, it can’t build rapport, read group dynamics, or respond to unexpected shifts in the discussion, making it unsuitable for moderating focus groups' nuanced and dynamic nature. 

Certainly, for more complicated qualitative formats or setups, e.g., deliberative events, presenting participants with balanced information about a product or issue and then facilitating structured discussions to explore their perspectives and preferences would take much work yet to deliver using AI. 

Equally, product and service co-creation, a collaborative process where businesses work closely with customers to jointly develop, design, and refine their offerings, will, for the medium term, require an experienced moderator to manage and deliver this process and experience.

AI will face significant limitations when attempting ethnographic market research, which relies on, certainly relative to other research methods, a deep understanding of culture and behavior through immersive observation. It is difficult to see how AI can replicate this in its present form. 

During ethnographic data collection, the best insight often arrives when the research participant forgets the researcher is there, or at least they are there to research them. 

Once this happens, the participant stops saying or showing the researcher what they think is socially acceptable and starts behaving naturally, speaking their minds. It is difficult to see how AI could capture this even in a more advanced state. 

Moreover, ethnographers must adapt to unforeseen, fluid situations during fieldwork, make subjective interpretations, and navigate relatively complex ethical considerations, such as informed consent and privacy boundaries. Even if AI was in a situation to gather this data,  AI is inherently structured and data-driven, meaning it will need help to handle ambiguity, subjectivity, and ethical dilemmas inherent in ethnographic research.

Finally, AI, even a more advanced form, will likely need more ability to comprehend the nuanced context of human interactions and cultural subtleties essential to ethnography. This method depends on establishing rapport, empathy, and emotional intelligence—qualities that AI, devoid of genuine emotions, will struggle to replicate effectively.  

While AI can assist with data processing and analysis at the end of the ethnographic research phase, I can’t see it replacing the human expertise, cultural sensitivity, and nuanced understanding necessary for meaningful ethnographic research, even in the longer term.

Tying together the threads of the three articles, I am (cautiously) hopeful. Indeed, market roles will change significantly over the coming years—the age of getting a junior researcher to code 500 open-ended responses to a questionnaire is almost certainly over. 

It's essential for us, as researchers, to adapt and stay relevant. 

Overall, we need to embrace AI as researchers. It will be crucial to understand its capabilities and limitations, where to apply specific tools, and where our role as researchers will remain. 

Human input into the market research process remains irreplaceable (for now). AI may excel at data processing but needs an intuitive understanding of human behavior, culture, and context to interpret meaning from research data. 

Researchers will continue to play a pivotal role in generating hypotheses, crafting research questions, and exploring unconventional angles that lead to more valuable insights.

Moreover, market research will require human expertise for strategic decision-making. 

While AI can analyze data and model responses, human researchers must interpret AI-generated insights, align them with business objectives, and support their implementation within organizations. 

The art of presenting and communicating research findings will continue to benefit from human involvement. The prospect of having a virtual talking head giving a PowerPoint deck over Zoom feels profoundly dystopian (an application of AI that I hope is a long way off!)

While AI will undoubtedly transform market research, human researchers' expertise, creativity, and nuanced understanding will remain invaluable. 

As we navigate AI’s nascent development, adapting, embracing, and enhancing our skills to remain relevant and deliver actionable insights that drive business success is crucial.