When Sharlene Seidman arrived at Johns Hopkins Medicine as the Vice President of Revenue Cycle, the revenue cycle team had been without a leader for a year to a year and a half. One of her first moves was not a dashboard review. It was a set of visuals showing the future structure, what functions depended on each other, and where silos had to come down. For revenue cycle leaders inheriting fragmented teams, that sequence is worth studying. The fastest way to miss the real problem is to manage rev cycle like a straight line of handoffs. In practice, it behaves more like a set of loops: access affects denials, coding affects billing, and breakdowns upstream come back downstream as rework.
Seidman has been at Hopkins five and a half months. What follows is not a finished transformation. It is the sequence a working rev cycle rebuild looks like when someone runs it well - and a rubric operators can apply to their own inheritance.
Before talking about vision, name the damage. A rev cycle team without a leader for a year to a year and a half is not simply "under-led." As Seidman put it, "that's tough for teams because they want to know that there's someone who has the vision, who's going to bring them along to a future." Cross-functional problems - the ones that sit between access and denials, or between coding and billing - tend to drift when no one has clear authority over them.
Her instinct was not to lead with a productivity mandate. It was to give the team something they had been missing: a picture of where this was going and who was responsible for what. She was direct about the ambition.
“We are good. We do things really, really well here,” said Seidman. “But we are going to take it to the next level and we’re going to become world class in revenue cycle.”
Seidman also noted that people first had to understand what "world-class" meant. Aspirational language only lands when it is translated into structure. That is why she paired the vision with visuals, showing which functions would be grouped, which would share a leader, and which dependencies would finally be reflected in the reporting lines. She said the reaction to those visuals was more than positive.
“(They said) I love what you’re doing,” Seidman said. “I really understand this. I think this is a great idea.”
That is not applause for a slogan. It is recognition of structure.
In Seidman's description, the redesign involved "some changes in positions and how we were thinking about what revenue functions depend on each other and should be grouped together under a single leader." That is the specific claim. It is not consolidation for its own sake. It is a call to look at where the loops actually run and put those loops under shared accountability.
For operators, the implication is a practical grouping test. When mapping the current org chart against the future one, ask four questions of each cluster of functions:
* Where does rework originate versus where is it absorbed? If registration errors surface as denials three weeks later, those two functions belong in the same accountability structure.
* Where is information shared continuously versus handed off once? Continuous-share pairs (eligibility and financial clearance, coding and CDI, denials and appeals) should not report through different leaders.
* Where does a handoff failure directly create a denial or a write-off? Handoffs with financial consequences need one owner, not two.
* Where can one leader reasonably own root cause? If the root cause of a denial category lives in access but the metric lives in follow-up, no one owns it end to end.
When dependent functions report through different leaders, the loops between them can become no one's clear responsibility. Grouping by dependency is one mechanism. Whatever the mechanism, the loops must have an owner.
Once the structure is drawn, the next question is whether the people in the boxes are the right ones. This is where diagnosis matters, because an underperforming manager is not always an unfixable one.
“(It can be) someone who wasn’t succeeding in their role, but you give them a new leader and they really explode and develop,” said Seidman
The operator lesson: performance issues are sometimes pairing issues. A manager failing under one leader may thrive under another. Someone whose skills look mismatched to their current group may fit an adjacent one once the structure changes.
Before replacing an underperforming leader, run three quick diagnostics:
Seidman is clear that some conversations do end in departures, and that "having those conversations takes some courage." The point is that the diagnosis has to precede the decision.
The related move: ask people what they actually want. Seidman notes that some employees are content where they are and others want to grow, but managers often don't know which is which until the high-potential person leaves for another employer. Adding a growth-aspiration question to every review is a retention control on the supervisor bench.
The other lever Seidman uses to build leadership capacity is the stretch assignment. Her example is specific.
About 10 or 11 years ago, her prior organization was trying to define patient financial experience and did not yet know what it was or how to make it better. Seidman used that ambiguity as a development laboratory: staff were pulled together outside their primary roles, asked to talk to other leaders and learn from other people, and expected to bring recommendations back.
The pattern generalizes to problems most rev cycle organizations still haven't cleanly owned: denials root cause across access and coding, estimate accuracy, prior authorization workflow redesign, self-pay conversion, or patient financial experience. Ambiguous, cross-functional problems with no natural owner in a siloed rev cycle are the kind of work that develops future directors while producing real answers. The wrong instinct is to hand them to a single department and ask for a plan. The right instinct is to charter a group, give them exposure to leaders they would not normally meet, and expect deliverables.
One caution Seidman raises: the managers of the best people will resist letting them stretch. In her framing, managers can become uncomfortable letting superstars expand because they fear losing them - but the leader's job is to help people grow so they stay within the organization, even if they move to different roles. Internal mobility is a retention strategy, not a threat to a headcount plan.
The loops conversation eventually runs into the technology conversation, because most of the friction inside the loops is administrative labor. Seidman is direct about what is coming.
“In order for us to survive, we are going to have to figure out ways to reduce the cost and the administrative burdens that we’re working with between payers and providers,” Seidman said.
She said future leaders will need to understand the technology "really well and how it can benefit all of us," and that AI fluency will be "table stakes." But she ties that fluency to a specific outcome: reducing cost and administrative burden between payers and providers.
The rubric that follows, framed as an operator framework rather than her direct words: evaluate every AI pitch against measurable operating math, not productivity claims. Useful measures to demand from a vendor or an internal build:
If a pitch cannot tie to at least two of these, the technology conversation has not earned the leader's time. This applies whether the capability is built internally, acquired through a vendor, or delivered through a managed service. The upstream requirement is the same: administrative cost has to come down, and the technology conversation only matters to the extent that it moves that number.
None of this works if the rev cycle leader treats the function as an internal machine. Seidman's framing on silos is worth quoting because it names the design principle beneath everything above.
“The silos are barriers and so the more and more we can work together and trust each other and depend on each other,” said Seidman. “You just become such a more powerful organization.”
In her description, rev cycle "might look like a series of handoffs but a lot of times you're circling back and there's so many integration points." That is the loops idea in Seidman's own words. And the loops do not stop at the edge of the rev cycle org chart.
Seidman calls this the "hundreds of bosses" problem. Everyone across the enterprise has a stake in whether the dollars come in cleanly and whether the patient's financial experience matches the clinical one. Her direct report group meets weekly, and while she does not run a formal program for cross-enterprise relationship building, she was clear: "we can't really do any of our work without getting to know some of our customers." In an academic medical center, those customers include service line VPs, the CFO's office, compliance, patient experience, and payer relations. If the internal loops are broken, the external ones will be too.
The takeaway is not that every health system should replicate Seidman's John Hopkins Medical build. The takeaway is that the capabilities she is putting in place - structural design that groups functions by dependency, leader-employee pairing diagnostics, stretch-assignment governance, AI evaluation tied to cost math, and cross-functional relationship discipline - are the ones a modern rev cycle needs access to, however they get acquired.
For a VP inheriting a fragmented operation, the near-term moves are concrete:
Seidman is five and a half months in. The rebuild is in progress, not finished. The sequence is portable. How each system acquires the capabilities - internal build, partnership, managed services, or a hybrid - belongs to the operator who knows the terrain.