How Synthetic Research Changed My Mind About The Mission of Market Research

By Matt Field

I’ve long been excited about what AI makes possible. Not just for us as consumers or business leaders, but for society at large. For example, GenAI is already driving better patient outcomes in Healthcare via earlier diagnoses, and broader access to knowledge in Education. Applied with care, AI can deliver incredible benefits for individuals and communities.

That’s why the chance to bring AI directly into our work at MakerSights, including through our new Synthetic Research product, has been so fulfilling. 

In this offering, we train custom LLMs to respond to common, high-value product prompts in minutes instead of weeks, at a fraction of the cost of traditional research. The “digital twins” we create are trained on real consumers, meaning their feedback mirrors their human counterparts with striking accuracy. It’s exciting to implement innovative and material ways to bring more affordable and timely consumer intelligence to our industry.

But I’d be lying if I said I didn’t feel hesitation at first.

Research has, at its core, always been about asking real people real questions. That’s what gives data credibility, especially in retail where so much time, money, and care is spent understanding and building authentic relationships with consumers. 

So when we first were exploring the idea of Synthetic Research, I had serious questions around whether an algorithm could ever match the value of going out into the community and engaging actual consumers. Did we risk sacrificing depth and nuance for speed and scale?

What I’ve come to believe deeply, as we’ve built out our Synthetic Research product over the last 18 months, and engaged hundreds in the retail community on the topic, is that it’s not an either/or choice. Synthetic data shouldn’t replace all organic (human) engagement. But it also cannot be ignored either. Not if you’re a brand (or research firm) truly looking to build a modern creation and GTM process that moves at the speed of today’s consumer. Here’s why.

Almost every brand aspires to be more consumer-centric, but traditional research makes that very hard. 

It’s by nature, not real-time: requiring surveys, briefs, and analysis. It’s also, by nature, expensive, with high inherit costs required for recruiting the right participants. For those reasons, it’s nearly impossible to scale across an entire seasonal assortment or marketing campaign. Even at our fastest and leanest, brands told us the same thing: “there’s still so much we’re not validating and testing with you (or anyone else) because our timelines are too tight and the cost is too high.” 

And that’s the key point. No matter how nuanced, rich or predictive traditional insights are, if they can’t be delivered quickly enough or within budget, then what’s the point? And with nearly every brand desperately seeking ways to shorten time to market, and do more with less, the pressure on organic research’s feasibility in modern retail will only grow. 

So here’s my encouragement on how to balance the two research approaches, ultimately achieving the benefits of both:

• When exploring new product technologies, category whitespace opportunities or brand perceptions, where timelines are less stressed and where the depth and context of feedback is critical - traditional human research remains the gold standard
• However, when you’re looking to make seasonal color, print or silhouette decisions across a wide range of SKUs, and in a very short timing window, then synthetic is the way to go. Fast, reliable and cost effective.   

Eighteen months ago, I approached Synthetic Research with equal parts optimism and doubt. Today, I see it more practically. When deployed correctly, it makes consumer intelligence faster, more accessible, and more actionable, turning research from a bottleneck into an advantage.

PS: You can learn more about Synthetic Research here.