For decades, commercial strategy in pharma has followed a familiar rhythm: forecast, launch, measure, learn. It has been an effective model for an industry where evidence accumulated over time and experience became the foundation for better decisions.
But the pace and complexity of pharmaceutical commercial planning have fundamentally changed.
Precision medicines, increasingly specialized patient populations, evolving payer dynamics, competitive launches, and more personalized models of patient engagement have transformed both the complexity of commercial planning and the cost of getting decisions wrong. Organizations are expected to make faster, more confident decisions, often before there is enough real-world evidence to eliminate uncertainty.
Every commercial strategy is ultimately an experiment. The question now is whether organizations should wait for the market to reveal the outcome. The next commercial advantage may not come from forecasting the future more accurately. It may come from testing strategic decisions before they ever reach the market.
From forecasting outcomes to simulating decisions
This is where the conversation shifts. Synthetic cohort simulation offers commercial organizations a different way to think about planning. Beyond forecasting what is most likely to happen, organizations can begin exploring how different strategic choices might perform before resources are committed in the real world.
At its core, synthetic cohort simulation combines real-world data, artificial intelligence, advanced analytics, and statistically representative virtual populations to create an environment where commercial strategies can be evaluated in silico. Rather than modelling the behaviour of individual patients, healthcare professionals, or payers, it enables organizations to understand how different commercial strategies perform across representative synthetic cohorts under different strategic scenarios.
Simulation explores how different strategic choices might perform by enabling organizations to test assumptions, compare alternatives, and explore multiple plausible scenarios before resources are committed.
The convergence of complexity and capability
The idea of using simulation to improve high-stakes decisions is not new. What has changed is its relevance – and increasingly, its feasibility – for pharmaceutical commercial organizations.
Over the past decade, advances in real-world data, artificial intelligence, commercial analytics, synthetic data, and digital twins have steadily expanded what pharmaceutical organizations can measure, model, and understand. Taken together, these advances are making simulation increasingly feasible as a form of commercial planning support.
This convergence comes at a pivotal moment for the industry. As precision medicines create more specialized patient populations and engagement becomes increasingly personalized, commercial teams are expected to make faster, higher-conviction decisions with less room for error.
Traditional analytics remain indispensable, but they were designed primarily to explain markets and forecast outcomes, not to rehearse alternative courses of action before strategic decisions are made. Synthetic cohort simulation complements these capabilities by creating an environment where commercial assumptions can be tested, strategic alternatives compared, and key decisions refined before resources are committed.
Simulation is already changing decision-making
Commercial synthetic cohort simulation is still emerging. The capabilities that underpin it are not. Across the pharmaceutical value chain, simulation is already reshaping how high-stakes decisions are made, first in research and development, and increasingly in areas that support commercial strategy.
Pharmaceutical R&D is already demonstrating what simulation can make possible. Synthetic control arms are becoming an accepted complement to traditional trial designs, while in silico simulation is increasingly being used to evaluate and refine trial designs before patients are enrolled. An AI-based digital twin methodology, PROCOVA™, which has been issued a positive Qualification Opinion by the European Medicines Agency, has demonstrated the potential for virtual patients to reduce trial enrolment while maintaining statistical power. QuantHealth has simulated more than 350 clinical trials across 23 therapeutic areas, reporting up to 90% predictive accuracy and $31.4 million in savings for one top-10 pharmaceutical company. Bristol Myers Squibb’s “Predict First” strategy reflects the same philosophy: using computational prediction to inform critical decisions before resources are committed.
Many of the same principles are now beginning to appear in commercial settings. Research from Stanford found that AI-generated agents can reproduce human survey responses with up to 85% accuracy, providing a technical foundation for new forms of commercial planning support. CVS Health is already using “agentic twins” based on more than 2.9 million consented interactions from over 400,000 individuals to evaluate medication adherence and care experiences. In testing, the synthetic population replicated known findings with up to 95% accuracy. Similar approaches are also beginning to appear in pharmaceutical market research, where synthetic respondents are used to explore engagement with rare patient populations before targeted validation.
Beyond technical feasibility, the commercial case is also beginning to emerge. Bain & Company reports that synthetic customers can halve research timelines, reduce costs to roughly one-third of traditional approaches alone, and enable continuous scenario testing. Research is already beginning to explore how LLM-based agentic simulation could support strategic rehearsal for pharmaceutical real-world evidence planning, enabling organizations to evaluate strategic choices before real-world implementation.
Together, these developments suggest that the foundations for commercial synthetic cohort simulation are already beginning to emerge. The opportunity now lies in bringing these capabilities together into a more coherent approach to commercial planning.
Learning before launch
The most significant shift is not the technology itself, but when organizations learn. Commercial teams have always refined strategy through execution and experience. Synthetic cohort simulation introduces the possibility of learning before execution.
Instead of advancing a single strategy into the market and learning from the outcome, commercial teams can evaluate multiple alternatives before committing resources – exploring messaging, market access assumptions, or engagement strategies across representative synthetic cohorts. The most promising approaches can then be validated through market research, real-world evidence, and expert review before broader deployment. Simulation does not replace human judgment, it helps ensure that judgment is focused where it creates the greatest value. Rather than asking experts to evaluate numerous possible alternatives, it enables them to focus on the options that have already demonstrated the greatest promise.
Over time, simulation becomes part of a continuous learning cycle. Simulation identifies the most promising approaches, market research and real-world evidence validate them, and those outcomes are then used to help refine future synthetic cohorts. Each cycle strengthens the next commercial decision. The organizations that gain the greatest advantage may be those that become better at exploring alternatives, validating assumptions, and rehearsing decisions before committing significant resources.
The next chapter for commercial intelligence
Synthetic cohort simulation is unlikely to emerge as a single technology or standalone platform. More likely, it will develop through the continued convergence of capabilities that are already reshaping pharmaceutical research, commercial planning, and decision-making.
Its greatest significance may ultimately lie less in the technology itself than in the way it changes how organizations approach commercial decisions. As commercial complexity continues to grow, synthetic cohort simulation may become an increasingly important part of how pharmaceutical organizations prepare for high-stakes decisions. Its greatest contribution may not be more accurate predictions, but better-informed choices, helping organizations test ideas, refine assumptions, and commit resources with greater confidence.
