AI in market research: Matt Gullett's predictions for 2024

by Matt Gullett | published on January, 2024

As we stand at the opening of 2024, it’s intriguing to consider the trajectory of AI in market research. The field is ripe with innovation, yet it's essential to navigate these technological winds with a cautious but adventurous spirit. Based on thoughtful analysis and industry pulse, here are my top predictions for AI's role in market research in the year ahead. 

Prediction No. 1: The double-edged sword of AI quality

Poor-quality AI outcomes will coexist with high-quality ones, continuing to emphasize the human element's importance in managing AI tools.

While AI has the potential to deliver astounding results in market research, it also possesses an equal proclivity for poor outputs. The disparity in quality doesn't stem from the technology itself, but from the manner in which we, as market researchers, harness it. Just as a novice sketch artist and a master painter can produce vastly different portraits using the same set of pencils, so too can market researchers generate contrasting outcomes with the same AI tools.

Let’s say a beverage company uses an AI tool to analyze customer feedback from social media. If the user fails to instruct the AI to look for brand-sensitive topics, or for potential PR-challenges, the AI may produce a valid list of insights without highlighting significant concerns present in a small yet important subset of the market. Contrarily, with a well-trained human in the loop fine-tuning the AI's understanding of language nuances, the outcome would be more reliable, revealing genuine consumer perceptions and not overlooking such risks.

Prediction No. 2: Traditional methods hold up against AI upstarts

Although a surge in AI-first market research methodologies will emerge, few will significantly outperform established research methods.

Innovation in market research is both a necessity and a given; it's the engine driving us forward. As we enter 2024, we can anticipate a wave of AI-first research methodologies claiming to redefine how we understand consumer behavior. However, among this surge, we should temper our expectations. While these novel methodologies shine with the luster of innovation, they must pass the critical test of improving our grasp of critical business issues. Most will likely result in an indifferent “meh,” providing insights that are no more insightful than those derived from time-honored techniques.

Let’s say a retail chain adopts a flashy AI-first methodology promising intricate consumer behavior insights solely from video analytics. Exciting at first glance, but upon closer inspection, they might find that this method overlooks the deep motivations and preferences that a more traditional survey or focus group might capture. In contrast, a more modest AI-enhancement to traditional methods could streamline processes and improve data robustness without compromising on depth.

Yet among the many attempts, a few gems will emerge. By adopting a fail-fast philosophy and setting clear benchmarks for success, researchers can quickly identify these viable advancements. The best improvements will likely come from methodologies that smartly augment traditional techniques, incorporating AI to handle greater volumes of data or introduce new variables into analysis without losing the tried-and-true insights that form the backbone of market research. The key is to maintain a balance: embracing innovation thoughtfully while holding fast to the reliable bedrock of proven methods.

Prediction No. 3: Spotlight on synthetic data

Synthetic data will undergo rigorous testing but will deliver mixed results, highlighting the need for meticulous implementation.

The narrative around synthetic data is becoming increasingly compelling as researchers look for ways to protect privacy while still extracting meaningful insights. 2024 will see a proliferation of tests with synthetic data, resulting in a mixed bag of outcomes. But the distinguishing factor between success and failure will lie in the quality of implementation. Synthetic data isn't just about creating a replica of original datasets; it's about capturing the underlying patterns and relationships while preserving the integrity and diversity of the original information.

Let’s say a consumer goods company, hoping to circumvent the limitations of sensitive sales data, decides to use synthetic data in testing market scenarios. If the synthetic data fails to adequately reflect the regional buying patterns due to inadequate modeling, the results could erroneously predict the success of a new product launch. This could lead to misallocated resources or lost opportunities. However, with a robust approach that ensures that key variables are accurately synthesized, the company could gain actionable insights into regional market dynamics, leading to strategies that are both effective and informed.

The lesson here is that synthetic data isn't a plug-and-play solution. It demands a methodical strategy, and its best-use cases will emerge from thoughtful experimentation. Firms that pair their synthetic datasets with human expertise — particularly in the realms of sampling and validation — stand the best chance of reaping the benefits without falling into the pitfalls of poor-quality implementations.

Prediction No. 4: Caution with custom AI

Industry-specific generative AIs will be developed, but researchers should approach them skeptically, acknowledging the risk of stifling the broad perspective essential to market research.

The market research industry will no doubt see an influx of generative AIs designed specifically for our unique needs. Take, for example, a bespoke AI built for the fast-moving consumer goods sector. It might be trained on vast quantities of historical purchasing data, survey results, and trend analyses. Initially, this specialized AI could impress by quickly generating reports that intuitively align with the longstanding patterns and recognized trends within the sector.

However, the caveat with such tailored AIs is their potential to circumscribe the broader perspective that is so crucial in understanding the complexities of human behavior. These AIs, while adept in their specific domain, may inadvertently overlook the innovative insights that can be gleaned from more diverse data sources. By training predominantly on siloed market research data, they risk missing out on the wider social, emotional, and cognitive factors that a more general AI might capture, thus limiting the depth and empathy of the findings.

The power of AI in market research is not just in reinforcing what we already know, but also in uncovering the previously unobserved interconnections that drive consumer behavior. It's in this broader view that we often find the most valuable insights. As we gravitate towards specialized AIs, let's ensure that we do not lose the breadth and richness of understanding that comes from engaging with diverse datasets and perspectives — the very essence that has always driven robust market research.

Bonus prediction: While AI can mimic creativity, the true innovative spark will remain human

As AI takes a more prominent role in the realm of creative endeavors within market research, its capabilities will spark a great deal of interest. The technologies will generate ideas that are seemingly fresh and original, offering convenient shortcuts to divergent thinking. For example, an agency using AI might produce a diverse array of product packaging designs in minutes, each blending colors, shapes, and textures in ways that initially seem groundbreaking.

While we will benefit greatly from this influx of AI-generated creativity, enriching our pool of options and accelerating the creative process, it's important to not discount the unmatched value of human innovation. The deep, groundbreaking insights and ideas that move industries forward spring from human experience, intuition, and imagination. Whereas AI draws from a well of existing patterns and concepts, the human mind can leap beyond the data, crafting the kind of trailblazing innovations that redefine markets.

Hence, as we welcome the creative power of AI in market research, let's also remember to nurture the human spark that complements this technology. It's the partnership between human insight and AI efficiency that will drive the most profound advancements. AI can push boundaries, but it's the human touch that pushes them in directions that are truly unprecedented and visionary.

Emerging prediction: Compliance becomes complex

Privacy and governance concerns will intensify, leading to increased involvement of corporate IT and risk management teams. Consider this scenario: A multinational corporation leverages AI for global market research but encounters different privacy standards across regions, similar to General Data Protection Regulation in Europe versus looser regulations elsewhere. Compliance becomes complex, and market research teams find themselves working alongside IT experts to ensure each AI application adheres to regional laws and best practices.

Adapting and thriving on the road ahead 

2024 promises advancements, lessons, and challenges for AI and market research. While we explore the potential housed within AI's algorithms, it's the human touch — our intuition, experience, and ability to see the larger picture — that will continue to guide how we use these tools for meaningful outcomes.

Each prediction I’ve offered underscores a facet of this partnership: the need for discerning oversight, valuing established methods while fostering innovation, the complex dance with synthetic data, the comprehensive perspective risked by overly narrow AIs, and the augmentation of human creativity with AI. All the while, privacy and governance loom as pivotal themes ensuring that our pursuit of knowledge is paired with responsibility.

As you consider these forecasts for AI and market research, reflect on how your organization can not only adapt to these trends but thrive within them. Positive outcomes require more than access to new tech; they demand a strategic, informed approach that marries the best of our human capabilities with the strength of artificial intelligence.

Matt Gullett, Bellomy’s SVP of Insights Technology, is a driving force behind Bellomy AI Analytics for Text. An employee of more than 20 years, he loves thinking and writing about AI.


Learn more about integrating AI into market research

At Bellomy, we pride ourselves on blending time-honored research methodologies with cutting-edge technology. Our staff and technologists possess deep research experience and are ready to help organizations harness AI responsibly and effectively. If you want to navigate the future of market research together, get in touch!