The Decision Enterprise: Why Health Plans Need a New Operating Model

Health plans have spent decades optimizing functions. Claims became faster. Utilization management became more sophisticated. Care management became more targeted. Risk adjustment became more rigorous. Quality became more measurable.

Yet the enterprise itself remains fragmented.

The same member can move through utilization management, care management, risk adjustment, quality, appeals, and customer service while each function works from a different representation of that member. The data may be accurate. The systems may be performing as designed. And the decisions can still be inconsistent.

That is no longer primarily a data problem.

It is an operating-model problem.

The next constraint is decision consistency

For years, the industry's technology agenda has centered on acquiring more data and connecting more systems. That work remains essential. But interoperability alone does not determine what an organization does with the information it receives.

A health plan can possess the same clinical evidence in multiple places and still interpret it differently. Different teams can apply different rules, use different terminology, work on different timelines, and reach different conclusions.

The result is an organization that is connected technically but fragmented operationally.

The next evolution is therefore not simply from less data to more data. It is from fragmented information to trusted, consistent decisions across the enterprise.

From data architecture to decision architecture

A Decision Enterprise begins with a different question.

Not: How do we get the data into another application?

But: How do we ensure that the enterprise can repeatedly make the right decision from the same trusted evidence?

That requires a connected architecture in which clinical evidence is acquired and normalized once, enriched with common business and clinical logic, and made available as reusable decision services across workflows.

Utilization management may use that evidence differently from risk adjustment. Care management may act on it differently from quality. Appeals and customer service may encounter it at completely different moments.

They do not need identical workflows.

They need a common foundation for deciding.

The Decision Enterprise

The model has seven layers:

1. Clinical Evidence — Capture the complete picture. Bring together clinical, claims, pharmacy, imaging, device, laboratory, and patient-generated information so decisions begin with sufficiently complete evidence.

2. Normalization & Interoperability — Make the data usable. Convert disparate information into a consistent, longitudinal representation that the enterprise can understand and reuse.

3. Business & Clinical Rules — Apply enterprise intelligence. Use shared terminology, policy, coding, mapping, and clinical logic so interpretation does not have to be recreated independently inside every function.

4. Shared Decision Services — Deliver decisions as a service. Expose reusable logic and evidence through services and APIs so applications and workflows can access the same trusted answers.

5. Operational Workflows — Embed decisions where work happens. Put those capabilities into utilization management, care management, risk adjustment, quality, appeals, customer service, and other operational processes.

6. Governance & Control — Operate with trust and accountability. Establish ownership, provenance, access controls, policy management, and oversight so decisions remain transparent, defensible, and governed.

7. Trusted Decisions — Consistently, across the enterprise. Deliver the right decision, at the right time, supported by trusted evidence—for every function and every member.

The important point is that these are not seven independent capabilities. Each layer depends on the one beneath it.

Why AI makes this more important, not less

AI changes the economics of interpretation and action. It can identify patterns, synthesize evidence, automate work, and dramatically increase the number of decisions an organization can make.

But greater decision velocity does not automatically create better decisions.

If AI is introduced into fragmented workflows operating from inconsistent evidence and logic, it can accelerate the fragmentation already present. The organization becomes faster without necessarily becoming more coherent.

The foundation therefore matters more as AI becomes more capable.

AI should amplify a trusted decision architecture—not compensate for the absence of one.

The shift that matters

The traditional health-plan operating model has been heavily claims-first. Claims remain indispensable, particularly as a financial and administrative record. But they are inherently backward-looking.

A more clinically informed enterprise can operate differently.

Clinical evidence allows the organization to understand what is happening closer to the point of care. Interoperability makes that evidence usable. Shared logic makes its interpretation consistent. Decision services make that intelligence reusable. Operational integration puts it into action.

The shift is therefore larger than interoperability.

It is a shift from information to insight, and from insight to coordinated action.

Better decisions become an enterprise capability

The value of this model ultimately does not reside in the architecture itself.

It appears in the outcomes.

More complete and defensible evidence can improve risk-adjustment accuracy. Shared services can reduce redundant work and manual chart chase. Common decision logic can create a more consistent member experience. Clinically informed workflows can improve quality and care outcomes. Governance can strengthen confidence in both human and AI-supported decisions.

And something more fundamental becomes possible:

The organization begins to learn as an enterprise rather than as a collection of functions.

Evidence is shared. Decisions become observable. Outcomes can be measured. Logic can be refined. What one part of the enterprise learns can improve decisions elsewhere.

That is the compounding advantage.

The next competitive advantage

Health plans do not need another layer of technology between their existing silos.

They need an operating model capable of turning the information they already possess—and the clinical evidence increasingly available to them—into trusted decisions at enterprise scale.

The next advantage in healthcare will not come simply from having more data.

It will come from making better, more consistent decisions—together.