The Exercise Of Human Market Research
Exercise is hard to do. The difficulty is the entire point of it. The difficulty of exercise is also what prevents most people from maintaining a consistent exercise regimen. It’s a vicious circle.
What if artificial intelligence could help us with this conundrum?
Imagine an office in the near future. A busy executive sits at his desk in the corner of the room, sending emails and sorting through spreadsheets all day. He’s accomplishing a great deal, professionally, but he’s been worried about the consequences for his health that will come from sitting at a desk all day long.
As luck would have it, a new AI company is providing solutions to address this problem. The executive orders himself a Run4UBot, and by the end of the week, he has the newest model up and running, literally, in his office. The product is a humanoid robot that runs on a treadmill all day long, accomplishing an impressive amount of physical exercise on behalf of the executive, so that he can focus on his job.
The Run4UBot exercises so its human owner doesn’t have to.
It’s a ridiculous idea, of course.
Buying a robot to perform imitations of physical exercise does not make the robot’s owner more physically fit. That’s because physical exercise is not a task that needs to be completed. It is an experience that people must struggle through in order to receive any benefit.
I contend that businesses that are replacing traditional market research methods with AI analytics are making the same mistake as an executive who buys a robot to perform physical exercises so that he doesn’t have to.
The Process Of Obtaining Insight
It’s common to think of market research in terms of its outcomes, to presume that market research is conducted in order to obtain useful information.
From this perspective, any aspect of a market research method that makes the process of obtaining information slow, expensive, or frustrating is a flaw that should be engineered away. Efficiency is the goal for people in business who think this way.
Using humans to conduct market research is slower, more expensive, and often more irritating than conducting an AI simulation of consumer behavior. These inefficiencies are causing many businesses to choose AI simulations over human market research.
What’s missing from such choices, however, is consideration of the process of market research as a benefit in itself.
Getting members of marketing teams out of their offices, away from their daily routines and corporate cultural influences, was a significant component of many market research processes, before the COVID-19 pandemic . Out in the field, market research clients encountered consumers who said and did unexpected things. They worked with researchers who asked questions that challenged internal assumptions.
Often, this process seemed unnecessary, or even felt painful. At times, clients would become exasperated, and just want it to stop, so that they could get to the answers they sought, the solutions at the end of the process.
A long-distance runner encounters similar frustrations, yet perseveres, understanding that pushing through the struggle is the only way to achieve the runner’s high, and to build the runner’s lean physique.
The frustrations of traditional, human-to-human market research methods were akin to the muscle ache experienced during exercise. These feelings hurt, but were also manifestations of minds being stretched, visions being expanded, and curiosities being provoked.
In-depth, face-to-face market research methods produced fertile ideas precisely because they were challenging ordeals that demanded their clients be mentally present through the process, rather than just showing up for a results meeting at the end. These research methods were rites of passage that provided insights as boons whose value was in proportion to the struggle of their clients.
For too many corporate cultures, the memory of these treks toward insight has faded, replaced by the instant glamour of AI dashboards. Electronic black boxes have removed all of the exertion from the market research process, and in doing so, have also removed its enduring rewards.
Aren’t Cognitive Tasks Different, Though?
The metaphor of market research as exercise could be criticized for its comparison of business to a biological experience. Market research, a skeptic might say, is not physical in the same way that a workout in a gym is.
Physical exercise is an embodied activity. It’s inherently biological, and focused on self-definition. The actions that we take, or choose not to take, with our own bodies, literally define our musculature. Exercise is a special kind of activity that is centered in our physical humanity. We can never outsource exercise to artificial intelligence systems, even if we implant AI within robots that physically imitate human bodies.
Could cognitive work be different from this? Might market research, because it deals with information, be effectively outsourced to artificial intelligence?
Certainly, there are some tasks within market research that could be performed by an artificial intelligence system without any consequences from the lack of human input. Quantitative analysis of standardized market data, for example, is often a matter of routine, going through a set of calculations that aren’t strengthened by a conscious mind doing the math.
Such routine sets of calculations, however, aren’t strengthened by the involvement of generative AI tools. In fact, generative AI models routinely get even simple mathematical calculations wrong because they don’t actually think. They merely imitate patterns in information, and are designed in such a way that they often fill in gaps in their performance with completely made up babble that looks plausible but is untethered from actual reality.
Besides, routine computational tasks exist within market research, but they don’t reach the level of the ultimate purpose of market research. In the end, market research isn’t about quantitative calculations. It’s about human behavior and motivation. A market research service that ends with a quantitative dashboard isn’t going the distance.
Effective, mature market research is a process of grappling with connections between complex ideas that develop in the interplay between human consciousness and a vibrant four-dimensional world that computers have never come close to expressing in quantitative form. Pattern-imitation algorithms based on the words that people type when they’re on the Internet just don’t cut it.
Generative AI analytic models used by market research firms dupe a lot of corporate executives because generative AI is good at imitating the language that people use when talking about market research. Generative AI models don’t think, however, and they aren’t good at gathering original data. So, generative AI as it’s currently used in market research is essentially a huckster that’s been trained how to imitate what market research looks like, without ever having learned how to go through the process of actual research. It’s like a college student who bluffs their way through class without having done any of the reading, covering for their ignorance by throwing a lot of big words around.
What if generative AI in market research was the genuine article, though? What if it actually learned how to do real research, and not just imitate the appearance of research?
Even if generative AI became capable of genuinely analyzing real world consumer motivation and behavior, market research firms would still lose out on the benefits of market research conducted by humans.
What happens when companies hire firms to do generative AI simulations of market research instead of doing real market research? Consider a modified version of the earlier story.
Imagine an office in the near future. A busy executive sits at his desk in the corner of the room, sending emails and sorting through spreadsheets all day. He’s getting a lot of tasks done, but he’s been worried about what it all means. What’s the point of it all?
As luck would have it, a new AI company is providing solutions to address this problem. The executive orders himself a Meditate4UBot, and by the end of the week, he has the newest model grappling with the issue. The product is a humanoid AI robot that sits in a meditative pose with its legs crossed all day long, analyzing the meaning of life from all the available data.
The AI robot meditates on behalf of the executive, so that he can focus on getting his tasks done. At the end of the week, the machine tells the executive the meaning of life, and how that meaning can be applied.
The Power Of The Process
Hiring an AI bot to meditate for you is ridiculous, of course. That is, however, what companies are doing when they pay for market research firms to perform imitations of market research using generative AI. Even if generative AI was capable of accomplishing research tasks, rather than just mimicking them, it would remain incapable of delivering the most valuable benefit of market research done by humans.
Market research isn’t just a task to get done. Effective market research is not just a technical process for obtaining information. It’s a social exercise designed to build cultural knowledge and foster social systems for applying human wisdom in practical, profitable ways.
Getting answers to research questions matters, but the real power of going through a market research project involving human beings is in the process. It’s an exercise that strengthens the minds of everyone involved and creates a shared experience that becomes a lasting social reference point enabling the client team to operate cohesively because they have grappled with ideas fundamental to their work, and they have been gone through that challenging experience together.
Market research strengthens client teams because it’s difficult. The struggle of a genuine market research process is cognitive and emotional, rather than physical, but it builds effective social concepts in the same way that physical exercise builds muscle. The capability that’s achieved through the process is enduring, and can be applied to projects outside of the literal subject that was the formal focus of research.
Cognitive Atrophy From Generative AI In Place Of Market Research
What happens when client teams skip genuine market research in favor of the quick and cheap alternative of Generative AI? There is not yet any scientific study that investigates this specific issue, but a recent study sponsored by Microsoft provides a pretty good idea of what the consequences will be.
The study focused on the cognitive impact of the use of generative AI by knowledge workers — a category that includes people who purchase and apply market research projects in the business world. The authors of the study conclude that use of confidence in the abilities of generative AI are associated with the erosion of critical thinking skills. They write:
“While GenAI can improve worker efficiency, it can inhibit critical engagement with work and can potentially lead to long-term overreliance on the tool and diminished skill for independent problem-solving. Higher confidence in GenAI’s ability to perform a task is related to less critical thinking effort. When using GenAI tools, the effort invested in critical thinking shifts from information gathering to information verification; from problem-solving to AI response integration; and from task execution to task stewardship.”
Microsoft is one of the dominant purveyors of generative AI. No one can accuse this study of being the product of outsiders taking cheap potshots at the AI industry.
Critical thinking skills are the foundation of effective decision-making in business. The best marketing teams are not data-driven. They’re thought-driven.
Generative AI models are undeniably impressive tools, but they don’t produce thoughts for us, and they discourage us from thinking. Business teams that rely on generative AI instead of genuine market research will find that their ability to think through the challenges of today’s chaotic economic and social conditions have become atrophied through disuse.
The consequences of replacing human market research with generative AI imitations becomes especially alarming when one considers the cultural function of market research within a corporation. Effective market research goes beyond mere data acquisition.
Worthwhile market research is designed as a tool for cultivating empathy between a brand and its consumers. In-depth market research is a strenuous process that requires the investment of time, effort, and mental focus. In return, however, market research clients gain a rich understanding not just of the ways that consumers think and behave, but also the ways that consumers feel. That emotional context can then be applied in ways that go far beyond the scope of specific research questions, and create a corporate culture that connects more authentically with customers, creating the elements that can eventually lead to enduring trust.
Market research is a form of exercise. As with physical exercise, there can be no shortcuts in market research. Generative AI that imitates market research is a tempting gimmick, but as with gimmicks designed to replace exercise, it doesn’t deliver healthy results.
It’s never too late to get back into the gym. The hardest part is getting started.
It’s time to put down the AI toy, get off the couch, and move again into the physical world, where your customers live and breathe.
Today can be the day you bring your brand back into human focus.
—
Jonathan Cook is a researcher with 30 years of experience conducting in-depth research into consumer culture and emotional motivation in the marketplace. If you’re looking for a more human approach to market research, get in touch.