A decade of Firely and FHIR: Lessons learned on the road to interoperability 

Last week, Firely turned ten—a milestone that calls for both pride and perspective. It’s not just a moment to celebrate what we’ve built as a company, but to reflect on the broader journey of FHIR itself. 

From the start, Firely’s story has been deeply tied to FHIR. We founded the company on a simple but powerful belief: that FHIR could—and should—play a central role in solving healthcare’s longstanding interoperability challenges, and ultimately, contribute to more affordable, patient-centered care.

My role, as I saw it, was to help turn that vision into reality—by transforming a promising standard into practical tools, services, and training, and by building a company around it with a sustainable business model. 

Now, as Firely celebrates its tenth anniversary and I begin my final six months as CEO, I want to share some reflections on the journey we’ve taken: where we’ve made progress, where we’re still stuck, and why I still believe in the potential of FHIR more than ever.

The FHIR paradox

When we started Firely, the business case for interoperability seemed like a no-brainer. Better data sharing means better decisions, better outcomes, and less waste. And yet, 50 years after the invention of the electronic medical record, we still don’t have interoperability by design. 

This is the FHIR paradox: the business case is easy, but the reality is hard.

FHIR is here to stay 

The good news? FHIR is no longer an experiment. It’s embedded in healthcare systems across the globe, backed by legislation, supported by a community of thousands. HL7 is more alive than ever, and FHIR version 6 is on the horizon as the first fully normative version. 

Firely, too, has come a long way. We’ve gone from a small startup to a true scale-up, from early prototypes to enterprise-grade solutions used across 30 countries. Our .NET SDK has been downloaded over 50 million times—and in the words of the King of Rock and Roll: “50,000,000 Elvis fans can’t be wrong.”

So why hasn’t it gone faster? 

Despite the momentum, healthcare interoperability remains frustratingly slow. Why? Because healthcare is different. 

In hundreds of conversations I’ve had over the last decade with customers, policy makers, and technologists around the world, I’ve come to see that healthcare interoperability is uniquely difficult for several reasons: 

• Healthcare is cautious, and rightfully so: Patient safety, data protection, and the ethical duty of care all slow the adoption of new technology. In healthcare, innovation isn’t just about efficiency; it’s about trust. You wouldn’t want your hospital to apply the rule of “move fast and break things”.
• The tragedy of diffusion: Those who feel the pain of poor interoperability—patients and clinicians—are rarely in control of solving it.
• Multi-dimensional complexity: Clinical, technical, organizational, and political complexity all intersect.
• Mismatch of incentives: Incumbents benefit from closed ecosystems and often resist open standards.
• Regulatory inertia: Governments alternate between bold mandates and bureaucratic delays.
• And most of all: it’s not sexy. Interoperability is essential, but invisible. FHIR is a background technology. A silent enabler.

What’s driving adoption? 

FHIR adoption has been anything but even. In my estimation, the U.S. is at least five years ahead—and the gap is growing. Ironically, the inefficiencies of the U.S. healthcare system create strong incentives for innovation. Combine that with assertive regulation—like the 21st Century Cures Act—and you get real traction. 

In Europe, things are more fragmented. While some countries are making progress, many mandates lack the clarity and enforcement needed to drive meaningful adoption. Germany’s ISiK policy, for example, has the right intentions, but lacks the detail and follow-through to drive real change. 

The pattern holds for use cases too. Adoption is still largely compliance-driven. Where regulation is strong and clear—like in hospitals and insurance—FHIR takes hold. Other sectors, like mental health, life sciences, and public health, are still catching up. 

The market landscape doesn’t always help either. Best-of-breed solutions have been squeezed out by larger vendors with less incentive to open up. And while U.S. cloud providers now offer FHIR APIs, they also introduce a new form of vendor lock-in: “cloud lock-in”. 

This is why the European Health Data Space (EHDS) matters so much—not just as a regulatory framework, but as an opportunity for Europe to lead by example, as it did with GDPR and the Digital Markets Act. 

Two forces shaping the future: AI and geopolitics 

Two forces are shaping the future of FHIR in ways we couldn’t have predicted 10 years ago. 

First, AI. Structured, computable data is essential for AI—and FHIR is built for that. Interestingly, we’re now seeing AI used to generate FHIR data, raising questions about whether the standard is a tool, a goal, or both. Either way, the relationship is mutually reinforcing. 

Second, geopolitics. The FHIR community has always been international, but rising tensions, regulatory swings, and funding uncertainty—especially in the U.S.—are straining that collaboration. 

However, this shift creates opportunities as well: Europe will have to do its share in accelerating the adoption of FHIR, rather than relying on the U.S. to lead the charge—similar to what we’re seeing with NATO. I firmly believe the EHDS cannot succeed without FHIR at its foundation, and not the other way around.

7 lessons learned 

After more than a decade in this space, here’s what I’d tell any business leader looking to implement FHIR: 

1. Start with a real use case: Don’t adopt FHIR just because it’s popular. Tech for tech’s sake rarely works. Focus on a real problem, and make sure interoperability—and FHIR—are the right fit.
2. Skip the PoCs: Especially in Europe, we see too many pilots that never go anywhere. If you’re building something, build it to go live—with real users and a plan to scale.
3. Don’t rely on subsidies alone: Funding can help, but it can’t carry a project with no business value. If it can’t survive without the grant, it won’t survive at all. Subsidies can be a curse in disguise.
4. Go greenfield first: Don’t modernize old systems for the sake of it—apply FHIR where it can offer the most value from day one.
5. Dare to go business-critical: Many organizations opt for low-risk pilots. But real impact comes when you use it in core, meaningful applications.
6. Understand build vs. buy: Open source isn’t free—it’s a build strategy. Use it if you have the skills and capacity; otherwise, supported solutions may save you a lot of time and money.
7. Know your stance on data: Is your data a strategic asset—or something locked away by vendors? FHIR raises this question. Be ready to answer it.

The road ahead

When we started with FHIR in 2012, I expected a revolution. In hindsight, I think we’re living through a tipping point. Not an overnight shift—but a gradual buildup to something transformative. 

FHIR APIs are becoming more common, richer, and more computable. We’re approaching the moment where the groundwork is solid enough that each new use case no longer feels like an uphill battle.

Instead, the path forward will begin to slope downward—easier, faster, and more scalable. Innovation will accelerate. And FHIR’s role will finally be what we always envisioned: the silent, enabling layer beneath a truly connected healthcare system. 

I won’t be in the driver’s seat for that part of the journey. But after 10 years of building Firely—and even more years believing in FHIR—I still believe this future is within reach. And Firely? We’re just getting started.