There is a pattern I have seen repeat itself across practices of every size. Leadership decides it is time to modernize the revenue cycle. A tool gets selected, a vendor gets contracted, and a rollout date gets set. Six months later, the tool is being used by maybe two people on the team, the rest of the staff has quietly returned to the spreadsheets and manual processes they started with, and the investment has not produced the results anyone hoped for.
This is not a technology failure. It is a change management failure, and it is far more common than anyone in the vendor community will tell you.
The good news is that it is also completely preventable, and the practices that get automation right tend to share a few specific habits that have nothing to do with which platform they chose.
That last number is worth sitting with. Eight full-time equivalents worth of time redirected away from manual claim status lookups and toward work that actually required human judgment.[4] That is not a small return. It is the kind of result that happens when automation is applied to the right problem with the right preparation behind it.
Why Most Automation Initiatives Stall
When I look at automation rollouts that have not gone well, the failure usually traces back to one of a few common patterns. The first is trying to automate too much, too fast. A practice decides to overhaul five workflows simultaneously, none of them are implemented cleanly, and the team ends up managing a half-working system while also trying to keep up with daily operations. The second is automating a broken process, which only produces broken results faster. The third, and most underestimated, is failing to bring the staff along.
"AI is not the barrier. Resources are. Teams need time, education, and support to make adoption sustainable."
Health system executive, HFMA / AKASA Gen AI in RCM Survey, 2025 [1]
This quote from a health system executive in the HFMA and AKASA survey captures something that gets overlooked in almost every automation conversation. The focus lands on the technology, and the people side gets treated as an afterthought. But the people side is where implementations live or die. Staff who feel like automation was done to them rather than with them will find ways to work around it, and usually will.
Research from Global Healthcare Resource reinforces this point directly, noting that automated RCM can trigger anxiety about job loss or role changes, and that addressing this proactively through clear communication and involving staff in workflow redesign is one of the most critical components of successful adoption.[5]
What Getting It Right Actually Looks Like
The practices that build lasting automation are the ones that treat it as a discipline rather than a project. They do not launch automation and move on. They build on it incrementally, measure it consistently, and keep refining it over time. Here is the framework I have seen work most reliably.
Start with the lowest-risk, highest-volume task
The best advice I have seen on this comes from FinThrive, a healthcare revenue technology company: start with automating the low-hanging fruit, meaning data-rich tasks that are manual, repetitive, and rule-based.[3] Do not automate to improve your claims process broadly. Identify one specific task within that process, such as scrubbing Medicare Advantage claims before submission, and automate that task well. Success here builds confidence and demonstrates value before you move to anything more complex.
Fix the process before you automate it
Automation amplifies whatever it touches. If the underlying process is flawed, the automated version will produce flawed results at higher volume. Before any tool goes live, map the workflow end to end and identify the gaps. What breaks under normal conditions? Where does information get lost between steps? Closing those gaps first is the work that makes automation actually deliver results.
Communicate what automation is changing and why
This step gets skipped more than any other. Before going live, the team needs to understand what is being automated, what their role looks like after the change, and why this makes their work better rather than their position vulnerable. The organizations that handle this well tend to involve staff in the tool selection and workflow redesign process itself, which builds ownership rather than resistance.
Measure performance from day one and adjust
Automation is not a set-it-and-forget-it solution. Define the specific metric you expect to improve before launch, track it from the first week, and build a regular review cadence into the process. If results are not materializing as expected, that data tells you where to look. The goal is continuous improvement, not a single implementation event.
The Traps to Avoid
Even well-intentioned automation efforts can go sideways for reasons that are entirely predictable once you have seen them enough times. These are the patterns worth watching for.
Poor integration with existing systems. Health systems consistently cite integration with existing platforms as one of the top barriers to automation.[4] A tool that cannot connect cleanly with your EHR or practice management system creates new manual work rather than eliminating it. Integration requirements should be evaluated before a contract is signed, not after.
Automating tasks that require human judgment. Not every RCM task is a candidate for automation. Effective automation targets objective-driven, rule-based activities. Tasks that require nuanced clinical reasoning, complex payer negotiation, or judgment calls about individual accounts still need human oversight. Misapplying automation to these areas creates more exceptions to manage, not fewer.
Scaling before the foundation is solid. The most common version of this I see is a practice expanding automation to new workflows before the first one is actually working reliably. Each addition to an unstable foundation adds complexity and compounds the existing problems. One workflow running cleanly is worth more than three running inconsistently.
A Real-World Scenario
Consider a primary care group that decided to automate eligibility verification after noticing that a significant share of their denials traced back to coverage errors at the front end. Rather than attempting to automate the entire intake process at once, they started with a single workflow: automated eligibility checks for all scheduled appointments, run 72 hours in advance.
Before launch, they spent two weeks mapping the current process, identifying where staff were doing manual lookups, and documenting the most common failure points. They also held a team meeting specifically to explain what the tool would and would not do, and to clarify that the role of front desk staff was shifting from manual lookups to exception review and resolution.
Within 60 days, front-end eligibility denials had dropped substantially, and the staff who had been resistant initially became some of the strongest advocates for expanding the automation to additional workflows, because they had seen it work and understood what they were being asked to do within it.
Where to Start This Week
Sustainable automation starts with clarity about what you are trying to solve and a realistic assessment of where your team is today. Here is a practical starting point for any practice working through this.
List the top three manual tasks consuming the most staff time in your revenue cycle. These are your automation candidates. Rank them by volume, repeatability, and risk. The one that is highest volume, most rule-based, and lowest risk is your starting point.
Map the process for that task before touching any technology. Document every step, every handoff, and every point where errors commonly occur. If the current process is not clean on paper, automation will not make it clean in practice.
Identify what metric you will use to measure success. Whether it is a reduction in eligibility denials, a decrease in manual staff hours on a specific task, or an improvement in first-pass resolution rate, define the measure before you start, not after.
Plan the staff communication before the tool launch. Who needs to know what is changing? What does their role look like after the change? How will they raise issues or exceptions when the automation does not handle something correctly? Answering these questions in advance is the difference between adoption and abandonment.
Automation that sticks is built one solid workflow at a time. The practices that get this right are not necessarily the ones with the biggest technology budgets. They are the ones that treat implementation as carefully as they treat tool selection, and that understand the humans running the system matter as much as the system itself.
In Issue #03, we will look at operational intelligence in practice, specifically how to use the data already sitting in your revenue cycle to make smarter decisions and surface the problems that are costing you money before they become a crisis.
- HFMA and AKASA (2025). Adoption of AI for hospital RCM surges, but cost, operational constraints slow progress. The Healthcare Financial Management Association (HFMA) is the leading professional organization for healthcare finance executives. AKASA is a health technology company specializing in AI-driven RCM solutions. This joint survey gathered responses from 519 CFOs and revenue cycle leaders at U.S. hospitals and health systems. Reported by Fierce Healthcare. fiercehealthcare.com →
- Becker's Hospital Review (2024). Debunking Three Myths in Healthcare RCM Automation. Becker's is one of the most widely read trade publications in hospital administration and revenue cycle management. This article provides a nuanced review of where RCM automation works well, where it does not, and what effective implementation actually requires. beckershospitalreview.com →
- FinThrive (2024). 3 Ways to Optimize Automation in Revenue Cycle Management. FinThrive is a healthcare revenue management company serving over 3,000 provider clients. This article draws on their operational experience implementing RCM automation at scale and provides practical guidance on where to start and how to scale effectively. finthrive.com →
- Fierce Healthcare (December 2025). Adoption of AI for hospital RCM surges. Fierce Healthcare is a leading healthcare industry news publication. The specific statistic referencing eight FTEs of freed staff time comes from a health system executive interview published in this article, describing real-world outcomes following RCM workflow automation. fiercehealthcare.com →
- Global Healthcare Resource (September 2025). Building RCM Teams in the Age of Automation. Global Healthcare Resource is a healthcare workforce solutions and consulting firm. This article focuses specifically on the human side of RCM automation, including change management, staff communication, and team redesign, areas that are frequently underaddressed in automation planning. globalhealthcareresource.com →
Ready to build workflows that actually last?
The ROI platform is designed around automation that is practical to implement and built to sustain results over time, not just at launch.