I attended the HBMA 2024 annual conference in Austin a couple of weeks ago, and the focus on technology was stronger than ever. Over the last 15+ years, I’ve never seen so many presentations and talks about automation and artificial intelligence (AI) in healthcare revenue cycle management.
While some of the presentations came from vendors who create and implement these technologies, many were from billing companies themselves. These companies shared their use cases and experiences, but there was a recurring theme of overstatement. Some billing companies seemed to exaggerate their technological advancements, and this became apparent through the vague and generic content presented. Often, there were more platitudes than detailed, actionable insights on the benefits of automation and AI.
Misconceptions About Medical Billing Automation and Artificial Intelligence
It was surprising how little some presenters distinguished between medical billing automation and AI. This confusion is concerning, especially from companies and individuals considered subject matter experts. To be credible on stage, a fundamental understanding of these two technologies is essential.
Besides hearing the presentations, I have personal insights into the actual state of these technologies within the industry. I’ve worked with many organizations in the HBMA network, served on technology committees, and have been involved in private equity acquisitions in this space. Having had a front-row seat to these companies’ operations, I’ve seen firsthand what’s really happening behind the scenes, beyond the claims made in public forums.
The Current State of Automation in Healthcare Revenue Cycle Management
One key takeaway is that automation in the medical billing industry is still in its infancy. The vast majority of billing companies have implemented little to no automation, and even fewer are adopting AI. The proportion of companies that have embraced significant automation is shockingly low.
There are many reasons for this lag, including the complexity of disparate systems that billing companies use. However, even among companies that should be leading the charge—those with hundreds of millions in revenue and large EBITDA margins—automation adoption remains minimal. Some companies do better in this area, often depending on the practice management system they use, but the general state of automation is underwhelming.
AI in Healthcare: Mostly Talk, Little Action
Artificial intelligence in this sector is even rarer than automation. In many cases, companies will include AI as a selling point in their investor offerings, but when you drill down, their AI initiatives are often small-scale and still in the testing phase. For example, some companies might be using machine learning to prioritize claims in a workflow, but this represents a tiny fraction of their overall operations.
Even among the largest billing companies, with quarter-billion-dollar valuations, the presence of AI is minimal. Yet, that doesn’t stop companies from discussing AI as if it’s a major part of their strategy. The reality is that AI adoption in healthcare revenue cycle management is still in its very early stages, and only a few companies have made even modest progress.
Why the Disconnect Between Talk and Action?
Why, then, is there so much talk about AI and automation if the industry is doing so little? Part of the answer lies in the dynamics of the industry. At conferences like HBMA, billing companies are talking to their peers, and there’s a lot of chest-thumping about technological advancements. This often comes from a place of insecurity—companies want to appear advanced, even when they know they have a long way to go.
A Path Forward: Learning Loops and Embracing a “Fail Fast” Culture
My recommendation is for companies to set aside ego and focus on learning. It’s okay to admit that there’s still a lot to figure out, that data is scarce, or that missteps will happen. The path to effective automation and AI is a long one, and companies need to develop the infrastructure and culture to support it.
A key part of this journey will be embracing a “fail fast” mindset, where companies try new things, test them quickly, and learn from their mistakes. This approach allows for faster iteration and improvement, helping the industry move toward real innovation over time.
By taking these steps, healthcare revenue cycle management companies can begin to make meaningful progress with automation and AI—beyond the buzzwords and into actionable, transformative change.
