Imagine if every important digital health care experience could be test driven first. You’d know what resonates before launching a new marketing campaign or how customer attitudes shift after different interactions. A new technology called generative agent simulations makes that possible.
Modeled on real individuals, these AI driven “agentic twins” behave like the people they represent and inform better customer experiences. It’s like giving every decision a rehearsal, so care journeys ultimately feel simpler and more intuitive from the start.
Generative agents model individuals, not averages
Over the past year, CVS Health has used generative agent simulations to help guide decision-making with the support of Simile, the first company to build and bring this technology to market.
Built on 2.9 million consented responses from more than 400,000 participants across 200+ behavioral scenarios, these agents are modeled on data from real people — including interview responses and past choices — so they act as safe, accurate stand-ins for the individuals they represent.
Unlike generic models that guess how an average person might behave, these agents reflect specific people and often hard-to-reach populations, letting us see how different groups respond to a message, workflow or service design.
Inside our digital test environment
We use generative agents to uncover friction across end-to-end user journeys, reach populations that are difficult to study quickly, test digital product changes before rollout and benchmark competitive perception across markets.
- Understanding the ‘why’ behind behavioral drivers. Simulations help us uncover the ‘why’ behind consumer preferences and motivations, revealing which experience factors matter most, for which patients, and when they influence behavior.
- Designing with speed and care. What used to require weeks of research now happens in hours. Teams can validate findings from previous studies and triage which new ideas deserve in-market pilots, speeding up how quickly we bring better experiences to patients.
- Accessing hard-to-reach populations through simulation. For health-critical or niche needs like chronic condition management, simulations let us explore questions that are otherwise too slow or sensitive to field directly, while keeping experts in the loop to interpret results.
Why patient experience leads to better health outcomes
We know that when patients have positive experiences, their health benefits. They’re more likely to take their medications as prescribed, have fewer acute medical events, and stay on track with vaccinations. The harder question to answer is why?
Take medication adherence, for example. To truly improve outcomes, we need to understand what shapes a patient’s willingness to stay on medication and why certain moments matter more than others. With Simile’s generative agents, we were able to condense weeks of research into hours, recreating patterns quickly to see what was driving decisions.
Our research surfaced clear priorities — trust in the pharmacy, confidence in medication handling, and convenience — and the reasons behind them. Follow up questions to simulated patients who weren’t adhering to their medication revealed barriers like confusion, anxiety about refill timing, and past frustrations.
This technology helps us move from knowing that experience matters to clearly seeing which interventions will have the biggest impact on patients.
From insight to real-world impact
Simulations don’t replace real-world research or human judgement. They help us prioritize what to test in market. We keep experts in the loop at every stage and maintain governance to monitor tone, fairness and safety so insights are actionable and trustworthy.
As we build from individual insights to bigger ‘what if’ scenarios across markets and communities, each step helps us design experiences that work in the real world and stay centered on the people we serve.
