The 17-Domain Map: Where Healthcare AI Actually Pays Off 

Every operational domain in healthcare has its own vendors, its own vocabulary, and its own case for change. Read enough of them back to back and a health system or health plan leader ends up with 17 separate pitches and no way to compare them on the same terms.

This page is the consolidated reference the book builds toward: one table, one set of criteria, so a CFO, COO, or CIO can see where the technology is mature, where the complexity is real, and where the return actually shows up, before deciding what to tackle first.

How to read this table

Four columns, four questions:

• Available Tech Maturity: how ready is the technology on the market today, not in a roadmap deck.

• Implementation Complexity: what it actually takes to stand this up, from integration lift to change management.

• Annual ROI (Large System): the order of magnitude a large health system or payer can expect once the domain is running well.

• Priority Tier: where this domain sits in a sequencing decision, not a ranking of importance.

The ROI figures assume a representative large health system: $3B in net patient revenue, 100,000 attributed lives, and 400,000 health plan members. Scale the numbers up or down from there for your own organization.

Domain Overview and Implementation Priority Matrix

Implementation sequencing, tier by tier

The table above sorts every domain by priority. The book's implementation sequencing goes a layer deeper: for each domain, why it sits where it does, and where a team should actually start.

Tier 1: Deploy in years 1 and 2. Highest ROI, most mature technology.

• Revenue Cycle Management: mature AI, an immediate and clear business case. $90-140M annually. Start with denial management and coding automation.

• Contact Center: fast to deploy, with a high automation rate achievable quickly. $12-28M annually. Start with routine screenings.

• Utilization Management / Prior Authorization: strong regulatory tailwinds from the CMS FHIR PA rule. $60-135M annually payer-side, $10-20M provider-side. Start with high-volume, low-complexity procedures.

• Claims Processing: the highest ceiling on the entire matrix, $190-425M for a payer, on the strength of very high tech maturity, though the implementation lift is the heaviest in this tier.

Tier 2: Deploy in years 2 and 3. High ROI, moderate complexity.

• Population Health Management: requires a strong data foundation first. The return shows up in care manager productivity and panel size. Deploy risk stratification and care gap automation first, outreach automation second.

• Care Coordination: builds on the same data foundation population health needs. The return shows up in patient safety, patient experience, and workforce satisfaction. ADT-triggered workflows are the fastest path to value.

• Pharmacy Operations: antimicrobial stewardship and order verification automation are already mature. Start there, then move to specialty drug management.

• Credentialing and Enrollment: a rapid implementation timeline that speeds onboarding. Start with roles that have high-volume turnover.

• Quality and Compliance, Infection Prevention: both carry value beyond the dollar figure on the table, a Star rating or a lower infection rate protects revenue and reputation in ways a single number understates.

Tier 3: Deploy in years 3 and 4. Strategic value, higher complexity.

• Laboratory Operations: requires real LIS integration expertise. Highest value for reference labs and large hospital labs.

• Supply Chain: requires ERP integration and procurement process redesign, with sustained value once it's deployed.

• Workforce Management: a complex optimization problem that needs culture change alongside the technical deployment.

• Imaging Operations: requires radiology leadership engagement. High value given procedure volume, start with scheduling optimization.

• Patient Support: sits here on complexity even though its ROI is domain-specific and meaningful in value-based contracts.

Tier 4: Deploy in years 4 and 5. Long-term and strategic.

• Clinical Trials: the highest ceiling in the book for a pharma sponsor or academic medical center, $50-200M, but complex integration requirements push it last. Start with trials that have high enrollment and simple eligibility requirements.

• Payer Operations (full scope): $80-120M from risk adjustment alone, but it requires payer-side architecture and is realistically suited only to integrated delivery and finance systems.

The pattern underneath the numbers

Look across all 17 rows and a pattern shows up that doesn't fit in any single column: the domains with the highest tech maturity and lowest complexity, revenue cycle, claims, contact center, credentialing, are also the domains where data already flows in structured, high-volume, rules-governed ways. The domains that lag, clinical trials, payer operations, workforce management, are the ones where the work is still fragmented across systems that were never built to talk to each other.

That's not a coincidence, and it's the same argument the book makes at the platform level. Autonomy doesn't arrive domain by domain because someone finally builds a good point solution for radiology or a good point solution for supply chain. It arrives when a health system's underlying data and workflow foundation is unified enough that the next domain doesn't have to start from zero. We enable autonomous healthcare operations, and this table is the closest thing to a map of what that looks like, function by function, across a real organization.

Where this table fits in the book

This matrix is one piece of a larger reference chapter. The book also includes a maturity self-assessment (score your organization's architecture and top domains from 1 to 5), a three-phase implementation checklist from data foundation through domain expansion, a governance charter template for the committee that oversees autonomous systems, and a full glossary of the terms used throughout. Each domain in the table above has its own dedicated appendix, A through Q, covering the technical architecture, the five levels of autonomy specific to that domain, and the detailed business case behind the ROI figure shown here.

Using this table

Three questions worth asking with this table in front of a leadership team:

• Which Tier 1 domain has the fewest internal blockers today, not just the best ROI on paper?

• Where does a Tier 2 or Tier 3 win build the data foundation a later domain will need?

• Which Tier 4 domain is worth starting now anyway, because the complexity is a people and process problem more than a technology one?