AI in revenue cycle: from point solutions to enterprise value

  • Blog
  • November 12, 2026

Key takeaways:

  • Redesign the work itself, not just a single step. Automating a single step leaves hand-offs and manual effort intact. PwC estimates coordinated, end-to-end redesign can produce upwards of 60% efficiency gains, 5% net revenue uplift, and 50% cost savings, though results depend on workflow selection, scale, and readiness.
  • Design for the roles AI creates. Today’s roles will fundamentally change as staff shift toward validating agent output, managing complex exceptions, and governing human-AI operations. Organizations that skip role-based planning will keep falling short on AI value.
  • Size the opportunity before choosing the tool. Pinpoint where labor, vendor spend, backlog, and leakage concentrate, and decide up front whether released capacity will be monetized, redeployed, or re-skilled.
  • Start with your EHR, and measure what it delivers. Use native capabilities where they meet or soon will meet your needs. Add third-party or custom tools where they clearly outperform or fill real gaps, and baseline each so results can be attributed.
  • Keep ownership after go-live. A Value Realization Office gives business, technology, and operations leaders shared accountability for tracking results against the business case and intervening when value falls short.

Extending PwC's broader exploration of AI's evolving role in healthcare, this paper applies that lens specifically to provider revenue cycle, examining how health systems can translate emerging AI capabilities into measurable operational and financial value.

AI is reshaping the revenue cycle

Across our work with health systems, we are seeing AI layered onto revenue cycle operations that still depend on manual handoffs, fragmented technology, and third-party labor. For example, an AI tool can draft a denial appeal in seconds. But if staff still need to locate the medical record, validate the letter, enter the payer portal, submit the appeal, and update the account, the workflow has not changed. Only one step became faster. That is the gap many providers are confronting.

The value of AI is marginal when applied to isolated tasks. Enterprise-level value emerges when organizations recognize AI adoption is an organizational transformation program, and redesign processes end-to-end translating automation capabilities into sustained capacity, productivity, and financial outcomes.

This gap is compounded by how AI is entering the market. EHR platforms are releasing new AI and automation features at a rapid pace. Members of PwC's Revenue Cycle Consortium have noted that while the pace of AI product development has created an abundance of tools, it has also created a prioritization challenge, as members navigate the difficulty isolating the ROI any single feature delivers once it's embedded in an actual workflow. As a result, many organizations struggle to articulate value from native EHR AI capabilities and to discern when to rely on native platform capabilities versus external solutions. That vacuum pushes them toward a third-party vendor ecosystem that offers something the native tools don't: a commercial-grade solution paired with a clearly defined business case. One built for fast governance, C-suite, and board approval. But this only compounds the underlying issue. Without a way to attribute results across native and third-party tools, providers still can't tell whether the value they were promised is the value they're getting.

PwC estimates suggest that coordinated AI-enabled redesign across targeted revenue cycle workflows can produce upwards of 60% efficiency gains, 5% net revenue uplift through reduced leakage, and 50% cost savings. These outcomes are not automatic. They depend on the organization’s technological landscape, the workflows selected, the scale of implementation, changing workforce requirements, organizational AI adoption readiness, and how released capacity is strategically redeployed.

Leading organizations are therefore focusing less on the number of AI tools deployed and more on how work gets done. They are redesigning workflows and determining whether the resulting capacity should be monetized, redeployed, or re-skilled.

Why aren't health systems seeing expected ROI from revenue cycle AI?

Across provider discussions, three patterns appear repeatedly.

1. The tool is chosen before the opportunity is sized

Organizations often begin with a visible pain point and select a solution before determining where labor, vendor spend, backlog, or revenue leakage is concentrated across the broader revenue cycle.

The business case may then focus on AI adoption metrics, such as transactions automated or users activated, rather than whether AI is driving performance improvement through cash acceleration, revenue uplift, vendor spend retired, or capacity released.

2. One task changes, but the workflow and workforce do not

AI can remove effort from an individual activity. However, when pre-existing handoffs, roles, exception paths, and decision rights remain intact, automation improves task-level efficiency, but leaves the broader operating model largely unchanged.

Op-model redesign is what converts task-level automation into enterprise value. Leaders must determine which activities can be eliminated, which can be automated, where human judgment remains necessary, and how exceptions should move through the process.

3. Ownership ends at go-live

Many implementations have clear ownership through selection and deployment, but no single function accountable for performance after launch. AI adoption requires a closed feedback loop between performance monitoring and intervention, empowering cross-functional teams to adjust the solution, workflow, or implementation approach when outcomes fall short of the expected value.

Organizations need defined ownership for:

  • Reviewing performance and human overrides
  • Refreshing payer rules and business logic
  • Adjusting confidence thresholds and exception paths, especially as operational needs evolve
  • Tracking financial results against the original business case
  • Ensuring scalability, while preventing compounding technical debt
  • Deciding how released capacity will be used

Without enterprise AI infrastructure and governance, performance plateaus and expected value remain difficult to prove.

How can health systems scale revenue cycle AI beyond point solutions?

Underpinning successful AI-enabled transformation is the need for a role-based view of how the revenue cycle operates today and will change moving forward. Today’s roles will fundamentally change as workflows are designed around AI. The skill sets and capabilities of tomorrow’s roles must evolve to support new workforce requirements, such as validating agent output, managing complex exceptions, and governing hybrid human-automated operating models. Without accounting for this critical variable, providers will continue to fall short on AI value.

1. Size the opportunity

Start by examining the automation potential of today’s revenue cycle roles and workflows, then identify where AI can create measurable value and where that impact can be clearly attributed.

For each workflow, assess:

  • Role-level impact (e.g. labor effort, manual touchpoints, task changes)
  • Vendor and point-solution spend
  • Backlog and unworked inventory
  • Revenue exposure (e.g. denial write-offs, underpayments, etc.)
  • Existing EHR capabilities and planned functionality
  • Automation potential and implementation complexity

Translate the analysis into a value heat map that identifies the functions with the strongest business case and establishes a baseline for value realization.

As EHR platforms continue to expand AI and automation capabilities, organizations need to distinguish between value created through native functionality and value created through custom or third-party agents. Without that baseline, it becomes increasingly difficult to determine what is actually driving improvement and whether the expected value is being realized.

The objective is not to automate everything, but to direct investment toward opportunities where the financial impact can be measured and captured with confidence.

2. Redesign the workflow

Redesign the process before deploying the technology.

Assess current workflows, remove unnecessary handoffs and manual steps, and determine where AI, automation, and human judgment should sit in the future state. Human intervention can then shift toward exceptions, payer escalation, and work requiring specialized judgment.

The question becomes less “Where can we add AI?” and more “How should this work be performed?”

Each future-state workflow should define:

  • What AI can execute autonomously
  • What requires human validation
  • What conditions trigger an exception
  • Who owns the exception
  • Which outcomes feed back into upstream processes
  • How human capabilities should evolve to manage agentic-driven processes

3. Assess the technology ecosystem and deploy repeatable Infrastructure

Before building new capabilities, assess what already exists across the EHR, revenue cycle platforms, clearinghouses, incumbent vendors, and emerging solutions. As EHR AI roadmaps continue to evolve, organizations should avoid creating custom capabilities that become difficult to unwind as native functionality matures.

Instead, organizations should:

  • Use existing capabilities where they meet business requirements
  • Prioritize native EHR functionality when it can address the need today or is expected to mature on the near-term roadmap
  • Buy or partner where the market offers stronger functionality
  • Customize where meaningful gaps remain
  • Retire overlapping tools that add cost without distinct value

The supporting infrastructure should allow these capabilities to work together through shared data, payer rules, workflow orchestration, access controls, and audit trails.

The goal is not a single platform that performs every task. It is an adaptable operating environment in which technological and human capabilities can evolve without disrupting how work needs to move across the revenue cycle.

4. Establish accountability for value

AI deployment needs clear ownership after go-live. A Value Realization Office (VRO) creates the governance structure to ensure AI initiatives continue to deliver against the business case, while identifying where proven infrastructure and capabilities can be scaled across additional use cases to extend value over time.

The VRO brings together business, technology, and operational leaders to:

  • Establish clear ownership and decision rights across RCM AI initiatives
  • Define how value will be measured and attributed to individual solutions
  • Track performance against the original business case and intervene when value is not materializing
  • Drive adoption through continual redesign of roles, workflows, and incentives around new capabilities
  • Govern the portfolio as EHR, third-party, and custom AI capabilities continue to evolve

Technology adoption alone does not create value. Organizations need a structured governance model that connects adoption and implementation decisions with the operational changes required to translate new capabilities into measurable outcomes.

When performance doesn’t deliver on the business case, the VRO should empower cross- functional teams to identify the source of the gap and take corrective action, ensuring that value realization extends well beyond go-live. Sustained returns depend on an ongoing management discipline that monitors outcomes, incorporates feedback from users and agent performance dashboards, and continuously adapts the technology and operating model as conditions change.

Executive takeaway

Proving that AI works is no longer enough. Health systems should demonstrate that AI delivers value. The imperative now is to scale what works and translate adoption into measurable outcomes. Your deployment approach will determine whether you derive the value AI promised.

Organizations that treat AI as a series of point solutions may improve individual workflows. Those that size the opportunity, redesign operations on a repeatable technology infrastructure supported via a formalized governance vehicle to ensure sustainable adoption will change revenue cycle economics.

Explore AI's role in healthcare

Discover agentic AI with AWS, Anthropic and PwC

Contact us

Scott Ramer

Scott Ramer

Partner, PwC US

Pete Johnson

Pete Johnson

Partner, PwC US

Shannon O’Shea

Shannon O’Shea

Principal, PwC US

Shakera Adenwala

Shakera Adenwala

Director, PwC US

Follow us