CEO Monthly Reflection March 2026: Synthetic Debate Getting Heated

By: Terence McCarron, CEO & Founder

Early this month, we held a company meeting to talk about Synthetic Sample.  It was time. 

We covered what “synthetic” means, how it’s built, what it needs, where it fits, along with different application use cases. Then, we had a candid conversation around where and if it fits in OpinionRoute’s offering.  We talked about the excitement and the obvious risks.  Where it may work and where it would be a major failure.

Now was the time because this has become a HOTLY debated topic in MRX.  Very thoughtful people who deserve our respect passionately advocate on opposite sides of the Synthetic sample debate. 

Researchers, academics, even brand-side researchers lament the insanity of the rapid rise of synthetic sample usage. The view that “Synthetic data is junk data” is widely held in this group.  They’re not shy about it either.

Meanwhile, other researchers, business leaders, and brand-side researchers all staunchly advocate with analytics that demonstrate the utility of synthetic in specific use cases.  “Synthetic is a powerful accelerator” or “future of research”. 

Webinars are popping up everywhere on the topic…  (By the way, those chat rooms have become the boxing ring of the synthetic debate.) 

In reflection, I find myself holding two conflicting thoughts. 

1. As a guy who fixates on sampling methodology, I continue to have unanswered questions around the Synthetic approach. 
2. As a business leader, I am excited about the novel technology behind it and the opportunity this presents to the MRX industry.

So much of the discourse feels like the transition I made in 2003.  I left phone research (and “Nat Rep” studies with probability sampling) to enter the world of Online Panels.  My friends from the AAPOR side of survey research hold firm that online sample is to be used with great caution, while acknowledging its popularity. One argument made the point long ago that “nonprobability samples have limited external validity .”

In 2026’s social media style, you might translate that line to “online sample is junk data”. 

So, what should we do about this? What’s the “right” answer?

To me, the answer lies in a process that I think has to happen. 

Much like the migration to online panels in the 2000s, there is an enormous appetite for synthetic data in our context today.  I also see a set of signals that end-clients do NOT want to explore this alone.  They are craving Insights agencies to help them figure this out.  It feels like an invitation to a journey in this new, maybe too early, and uncertain area.

As a business owner seeing those signals, I’m inclined to take that journey with my clients.  As I do, I bring my experience and unanswered questions as invaluable resources as we focus on helping the client succeed.  The more we try, the more we share, the faster we’ll either have a breakthrough or a breakdown of the idea. 

Want to talk about synthetic sample experimentation?  Contact us here!