Breaking Free from Pilot Purgatory: A Path Forward for Healthcare Innovation

As Chief Medical Officer of West Health, an organization that is dedicated to lowering the cost of healthcare and advancing sustainable, high-quality delivery models in healthcare, I have witnessed a troubling pattern repeat itself countless times. A promising clinical initiative launches with enthusiasm and early success, only to languish later. The results look good. The team is energized. Leadership nods approvingly. But then… nothing happens. The pilot becomes a perpetual experience, isolated from the rest of the organization, while the broader system continues unchanged.

Zia Agha, MD

What is the “death by pilot” phenomenon? 

This phenomenon has earned the name, “death by pilot”. The reasons are complex and interconnected: Research has identified that securing sustained commitment across multiple stakeholder groups, managing unanticipated technical problems, and defining appropriate scope are three major challenges that lead to pilot failures.

Beneath these challenges, however, there is a deeper structural issue: we have been approaching innovation without considering its ability to scale across the organization.

Here lies the irony: pilots often fail because they succeed. When a small team implements a new technology for 50 patients on one floor, they can customize everything. They can configure systems, get technical support, and adapt on the fly. The pilot proves that the concept works and everyone involved feels the excitement of making a real difference.  

But scaling that same success to 5000 patients across 10 facilities?  That’s where the dream collides with reality. The infrastructure that made the pilot possible, such as the dedicated team, the manual workarounds and customizations become unsustainable at scale. What once worked through close coordination and individual effort now strains under the weight of volume, variability, and competing priorities. The custom integrations that connected seamlessly to one facility’s systems encounter different configurations, different versions, and different constraints at each new location.    

When the pilot stalls, it often looks like a commitment problem, but more often than not, it is a capacity problem.

The staff who championed the pilot are now stretched thin, balancing full clinical duties, compliance tasks, and competing responsibilities. Without additional support or infrastructure, the work that once felt innovative becomes unsustainable. Momentum fades, data collection slows, and the processes that held the pilot together begin to unravel. The system simply hasn’t evolved to carry the weight of its own success.

So what is the solution?

It isn’t to stop piloting. Pilots serve a necessary purpose to validate ideas and build stakeholder confidence. The solution isn’t to demand more organizational commitment either. Commitment alone can’t overcome structural barriers. Instead of isolated pilots followed by attempted scaling, we should embrace phased implementation at scale from day one.  

Phased implementation at scale means designing for enterprise-wide adoption from the start while rolling out in deliberate stages. The critical difference here is intention: every phase is built as a permanent component of the eventual full deployment, not as a separate experiment, requiring translation later.

Begin by asking different questions 

The approach begins with defining the end state. What if, before launching any pilot, we asked different questions?  What if we asked, “what does this innovation look like when fully implemented across all relevant sites, departments, and populations?”, “what infrastructure, training, governance, and resources will it require?” By answering these questions upfront, we ensure that each phase builds directly toward our ultimate goal.  

When we design a pilot, we optimize for proving efficacy under ideal conditions. When we design for phased scale, we optimize for sustainable integration into messy reality. We identify the training programs that must exist permanently, not just for early adopters. We build the technical infrastructure that can handle volume, not just pilot numbers. And we secure the organizational commitments and resources required for full implementation before we begin, rather than discovering halfway through that scaling is impossible without them. Phased for a scaled implementation takes front-loaded planning, which can feel more cumbersome than launching a quick pilot, but it dramatically increases the likelihood that the innovation will actually reach the patients who need it most.  

This shift in perspective leads to innovation that is built on scalability from the start. It’s about thinking through technologies and how they can integrate seamlessly with existing systems and solutions that work within current workflows, leveraging the staff and infrastructure already in place rather than introducing an entirely new resource.   

Healthcare organizations already have extraordinary resources: talented staff, sophisticated technology systems, and deep operational knowledge. Healthcare innovations should leverage what is already there. Yet despite these assets, studies show that limited funding for scaling and regulatory hurdles often prevent promising AI solutions from moving beyond experimental states to full-scale implementation. Rather than building entirely new systems from scratch, successful innovations focus on integrating with existing workflows, aligning with staff capacity, and addressing real clinical needs. By meeting teams where they are, by using the tools, processes, and expertise already in place, healthcare organizations can move away from “pilot purgatory” to scalable, lasting solutions.  

The real measure of innovation isn’t the brilliance of the idea. It’s whether it can survive the transition from small-scale experiments to everyday practice. This is what separates fleeting pilots from lasting change. Healthcare, with all its complexities and challenges, requires innovations that fit into real-world workflows, align with existing resources, and can be sustained to make a meaningful difference.