Q&A with 2025 Health Care/Pharmaceutical Research Project award winners, Vital Findings and Dexcom
Editor’s note: Vital Findings and Dexcom are the winners of the 2025 Health Care/Pharmaceutical Research Project award, a category in the Marketing Research and Insight Excellence Awards. The award winners were announced during a virtual celebration on November 19, 2025. To learn more about the awards, visit https://www.quirksawards.com/.
Vital Findings and Dexcom won the 2025 Health Care/Pharmaceutical Research Project award for a breakthrough segmentation study that resulted in the creation of Stelo. The category, part of the Marketing Research and Insight Excellence Awards, acknowledges a marketing research-related health care or pharma research project that took an innovative approach.
The two organizations built the segmentation, then used AI to bring it to life at speed and scale. The result was vivid, photorealistic personas that were used in AI videos to introduce each segment. Custom chatbots let teams “talk” to these segments directly.
The result was a confident, fast launch of Dexcom’s new product, Stelo.
Jason Kramer, chief research officer at Vital Findings, said he is thankful to the Dexcom Stelo marketing team for being open to this experimental approach.
“They were willing to experiment with AI, and the product now has AI features built in, so they've continued to embrace it,” Kramer said.
Describe the project and the impact it had on Dexcom.
Dexcom partnered with Vital Findings to develop a breakthrough segmentation for Stelo, the first continuous glucose monitor available without a prescription. We had time to conduct a robust quantitative segmentation. But with Stelo’s launch only a short time away, there wasn’t time or budget for deep qualitative research. We needed a bold, fast solution.
So, we turned to AI.
Using AI, we generated media usage profiles for each segment, created photorealistic imagery, produced AI-powered videos and even built interactive chatbots so teams could “talk” to each segment.
The result?
Dexcom executives called the rollout “mind blowing.” The marketing and media teams quickly aligned on the priority segment, helping drive a targeted and confident launch strategy.
The AI approach was so successful it is now being expanded to EMEA.
How do you see AI impacting the way researchers conduct segmentation studies over the next two years?
AI is already becoming an essential tool for segmentation. We use it to develop hypotheses of which types of segments might be in the market. We use AI-powered open-ends to give us richer qual feedback in quant. We use qual-at-scale/AI moderation to bring segments to life faster and to give us the deep, value- and motivation-based data we need to create rich personas.
But the biggest shift is our ability to keep the segments relevant. With AI persona chatbots, we can literally put the segment in the room for key discussions. Clients still validate before making big decisions, but for a quick directional answer, there's never been a better tool.
Please share one or two key takeaways from this project.
At the time, we were really pioneering AI image and video generation tools, and they've since evolved spectacularly. But what we were able to achieve at the time taught us that we didn't have to settle for just adding quotes to a PowerPoint slide – we could create imagery in situations where we couldn't get real pictures or video. From an ethical standpoint, we're careful to label everything as AI. But the visuals are leagues beyond text in giving marketers a feel for the segments.
Another key learning was about creating good chatbots/AI personas. Sometimes people think you can just dump your data into a platform and ask it to do what you want. The reality is that everything needs to be distilled, and you need strong analysis going in for the outputs to be strong.