AI In Market Research: The Story So Far – Chapter 1: Adapt or Get Left Behind

Whether you like it or not, AI is here to stay.  Yes, AI is a threat to most jobs, including those in the market research industry, since it shortcuts processes while optimizing operational efficiency.  While market research technology didn’t develop as fast as other industries in the early to mid-2000s, the advent and subsequent mainstream appeal of AI has forced market research to get with the times.  You’re in trouble if you fail to embrace it but if you do, you get to be on the winning side.

Experts expressed that we’re still in the early exploratory stages of AI but there is already depth in its application in market research. Take, for example, the humanization of surveys.  An interactive and dynamically probing AI improved overall data quality in more than one experiment due to an increased engagement from respondents resulting from a sense of appreciation over the perceived but simulated attention paid to them and their responses during the survey.  In the same vein, employing a conversational AI voice has been shown to dramatically drive engagement for better data.

That latter effort to humanize surveys has created an influx of voice responses and content, leading to the new question of what we should now do with all those resources, which would be a byproduct of AI-based solutions.  Of course, LLMs and other existing AI models would be employed to help find the answer to this question. 

Aside from solving dark data, AI has also displayed impressive capabilities to answer choice tasks, especially performing well with well-known topics and products, even outdoing humans in some surveys where humans get confused or find it hard to render a judgment.  It’s also been considered for AI to adapt existing survey data for a new topic to save time.  AI’s role in market research might still be experimental at this point, but it has grown to the point where it’s being utilized and adapted to take on one challenge after another.

Our second entry in this four-part blog series highlights some of AI’s risks and limitations, and how understanding and mitigating these factors can lead to their effective and optimal utilization.