Black Book: AI may make some outsourced health plan work cheaper to bring in-house
U.S. health plan administrative costs rose 2.7% to about $69.2 billion in the first half of 2026, even as margins improved to 4.1%. Black Book Research's 2027 outlook says AI is starting to move payer sourcing in both directions. Demand for specialist vendors stays strong in some functions, while automation could make in-house or hybrid models viable in others. The report recommends comparing each major outsourced service with rival vendors and the best internal or hybrid option, and checking whether projected savings become actual cash.
This story was produced through MarketScale. See how Healthcare teams put it to work with Executive Thought Leadership.
Key takeaways
Every major outsourced payer service should be benchmarked against a rival vendor and against the plan's best feasible internal or hybrid model, a three-way contest at each renewal.
Avoided vacancies, projected productivity gains and payment recoveries are not recurring cash, so outsourcing savings should not be assumed to fund AI until the service, AI and funding cases each stand on their own.
Provider data sits on both sides of the sourcing question: it is a top pain point and a specialist-demand area, yet also one of the functions where automation could make hybrid or in-house models viable.
Free workspace
Turn your Healthcare expertise into content.
Record interviews, organize footage, and write with AI on a free trial of the MarketScale platform for qualifying companies. No demo required, no credit card.
Across the U.S. health insurance industry, net income climbed to about $27.7 billion and profit margins improved from roughly 1.8% to 4.1%. The bill for adjusting claims and running administration went up anyway, rising 2.7% to approximately $69.2 billion in the first half of 2026.
Black Book Research highlights that mismatch in its newly released Payer, Health Plan & Managed Care Outsourcing Outlook 2027. According to the report, industry profitability among U.S. health plans has strengthened heading into 2027, yet administrative costs remain a stubborn problem. Black Book says that gap is pushing payer executives to test whether outsourcing and AI deliver genuine operating savings that recur year after year, or simply move expenses somewhere else.
The answer, according to the report, differs by function. AI is pushing some payer work toward specialist vendors and making other work cheap enough to bring home. For operators facing a contract renewal, that points to the report's recommendation: benchmark each major outsourced service against rival vendors, and also against the best feasible in-house or hybrid alternative.
Savings that never reach the bank
Black Book's sharpest warning is about accounting. The report tells plans not to count four kinds of projected gains as recurring cash savings by default: staffing reductions, payment recoveries, avoided vacancies and productivity improvements. Its concern is productivity gains that never become available cash.
That caution matters most for plans counting on outsourcing to help fund AI. Black Book found that some health plans are turning to outsourcing and managed services to free up operating dollars for AI investment. Others are buying AI-enabled services for claims, utilization management, provider data and member operations. Both routes rest on the assumption that a savings figure in a business case will show up as money.
So the report recommends splitting one decision into three: a service case, an AI investment case and a funding case, each evaluated on its own before anyone concludes that outsourcing savings can bankroll new technology. It is a dull-sounding discipline with real consequences. A deal can be a sound service decision and still fail as a funding source.
The industry's trade association frames the same gap from the technology side. AHIP's description of a webinar with leaders from Accenture and CareFirst BlueCross BlueShield, including Accenture managing director Terry Hemken and CareFirst chief digital information officer Dori Henderson, says many payers have not moved past pilots and proofs of concept, and lists unclear value measurement alongside fragmented data, limited governance and difficulty scaling as common barriers. Black Book's three-case test is one way to make the value measurable before the contract is signed.
Before an outsourcing deal is cited as the funding source for an AI project, ask finance which line in next year's actual budget gets smaller, and by how much. If the answer is an avoided hire or a productivity estimate, the funding case isn't made yet.
Which functions lean out and which lean in
Black Book found no single sourcing trend across the payer enterprise. The survey behind the report, drawn from its 2025-2026 payer technology program with 8,194 verified respondents across 60 technology and service domains, shows where pressure concentrates.
Black Book sees continued strong demand for specialist partners across claims operations, cybersecurity, AI-enabled workflows, risk adjustment, data and interoperability, quality, provider data, utilization management and regulatory operations. On the other side of the ledger, the report identifies provider data, credentialing, claims administration and selected administrative workflows as functions where automation could make hybrid or internally managed models more economically viable, because the technology reduces the labor that made outsourcing attractive in the first place.
Provider data appears on both lists. With 79% of respondents reporting provider-data problems, that overlap suggests the right answer depends on the plan: one with clean source systems might automate in-house, while another might need a specialist to fix the data before anything can be automated.
Vendors are now building automation directly into the services they sell. Wipro says its PayerAI model has three prongs. The first is a payer-in-a-box component that lets plans plug modular solutions into the systems they already run. It improves work like eligibility verification, retention or back-office administration for Medicare, Medicaid, ACA and commercial plans. The second, Cognitive Claims Lifecycle Management, applies smart adjudication and actionable insights to streamline claims. The third, Agentic AI Customer Services, is meant to transform support through AI-powered agent assist, real-time translation and voice neutralization.
Black Book's advice turns every outsourcing renewal into a three-way contest between the incumbent, a rival vendor and the plan's own best internal or hybrid option. The report recommends running that comparison periodically for every major outsourced service, not only when a contract lapses.
None of this adds up to an insourcing wave, and the report does not claim one. Black Book founder Douglas Brown described 2027 as an operating-model reassessment rather than an outsourcing expansion cycle, and the long list of functions where specialist demand remains strong points the other way for much of the enterprise. A plan could end the year with more outsourcing in utilization management and less in credentialing.
The 2027 API clock
The reassessment runs straight into federal deadlines. Starting in 2027, plans must meet broader requirements for interoperability and prior authorization APIs, and electronic claims-attachment rules move toward mandatory implementation in 2028.
The 2027 requirements Black Book cites cover four API areas: Patient Access, Provider Access, Payer-to-Payer exchange and prior authorization. Electronic claims attachments follow on a 2028 track.
Black Book argues these dates make it hard to separate sourcing from technology, compliance and data strategy. Brown's framing is blunt: a plan can hand a process to a vendor, but the regulatory accountability stays with the plan. With prior authorization the top priority for 86% of respondents and interoperability cited by 82%, the functions under the most regulatory pressure are also the ones where outsourcing decisions are most active.
For plans that outsource prior authorization or data exchange work, that suggests settling data access and oversight terms before signing rather than during implementation.
Who does the work, where, and on whose cloud
The report draws a line that sourcing debates often blur. Outsourcing decides who performs the work. Offshoring and nearshoring decide where. Black Book recommends that plans judge delivery models on more than geography alone, weighing cybersecurity, data-access controls, staffing, business continuity, subcontractor dependencies and operational performance.
It also flags a quieter risk: vendor diversification that only looks diversified. Several suppliers can depend on the same cloud infrastructure, so a plan that spread claims and member services across different vendors may still have a single point of failure underneath.
The calendar sets the next test. The expanded API requirements begin in 2027, and electronic claims-attachment requirements move toward mandatory implementation in 2028. For plans making sourcing choices now, that could mean checking whether the automation they insource, or the vendor they keep, can absorb a second mandate without a new contract.
Sources
- Payer Outsourcing Enters a Reset as AI Rewrites Health Plan ... ↗
- Payer Outsourcing Enters a Reset as AI Rewrites Health Plan Economics ... ↗
- Reinventing the Health Care Payer Enterprise: Turning AI… - AHIP ↗
- PayerAI: AI-Driven Healthcare Payer Solutions for Health Plans - Wipro ↗
- Register for the Payer, Health Plan & Managed Care Outsourcing 2027 Outlook Report- Black Book Research ↗
Featured companies
Your experts belong here
Every story in MarketScale Healthcare starts with a company putting its clinicians, service-line leaders, and field engineers on the record. Buyers are already reading this topic. The only question is whose experts they find.
Service-line buyers vet vendors quietly, and your clinicians become the proof they find while doing it.
About the author
The MarketScale Newsroom reports on the companies, technologies, and trends shaping 16 B2B industries. It turns primary sources and expert commentary into clear, useful coverage for the people doing the work.