Enterprise Asset

Operational Playbook: Care delivery must adopt AI strategically to survive the 21st century

Download Research

This operational playbook is for care delivery enterprise (health systems, hospitals, and ASCs) CIOs to leverage technology, particularly AI, to address four key existential challenges by their organizations: 1) revenue headwinds, 2) margin deterioration, 3) clinician shortages, and 4) patient expectations for speed, accuracy, and costs. It centers on AI as the core technology enabled by a new delivery paradigm framed by Services-as-Software™ (SaS).

Addressing these challenges with the specific steps in this playbook will yield tangible, quantifiable benefits such as increased revenue, reduced costs, improved clinician productivity, and higher patient engagement rates.

HFS Operational Playbooks are practical guides to solving key enterprise challenges that consume significant costs, time, and resources. The playbook provides enterprise leaders a realistic roadmap with specific “to-dos” to address their everyday challenges so they can clear mental and financial space to deliver next-level value.

The problem: Care delivery enterprises are facing an existential crisis at the nexus of multiple, compounding challenges

The US healthcare delivery landscape, despite the enormity of AI-induced opportunities, is the most economically challenged industry segment. Externally driven headwinds include the One Big Beautiful Bill Act (OBBBA), which reduced Medicaid funding by close to 10% annually, slashed reimbursement rates across the board (primarily for Medicare), and a rapidly aging population with a high prevalence of chronic conditions that will continue to drive up the cost of care. Internally driven challenges include slow and low technology investments that prevent higher productivity and efficiency, unrealized benefits from M&A activities, and slow progress in finding new sources of revenue.

These challenges will manifest into four categories of threats that need immediate remediation. The rationale for urgency lies in the bleak data from the Cecil G. Sheps Center for Health Services Research at UNC, which identified 338 rural hospitals as particularly vulnerable. Other data points suggest that even urban and inner-city hospitals are vulnerable and at risk for closure.

Revenue headwinds

Shrinking reimbursement rates by the Centers of Medicare and Medicaid (CMS), exacerbated by site-neutral payments, increased payer friction seen in higher rates of denied or delayed payments, prior authorization, and increased uninsured rates, are reducing the number of paying patients.

These factors will reduce revenues by 3% to 6% over a 12-month period and by 7% to 14% over a three-year period for those that survive this tsunami of challenges.

A two-panel infographic illustrating the revenue headwind threat. The left panel is a line chart titled "Reimbursement rates could shrink by 14% in three years," with the vertical axis scaled from 80% to 100% and the horizontal axis showing four time points: Now, In 1 year, In 2 years, and In 3 years. Two lines are plotted: a best-case line (purple) that declines modestly, and a worst-case line (orange/red) that falls more steeply to approximately 86% by year three. The right panel is a directional table showing factors driving the revenue challenge: Demand for care (up arrow), Population age (up arrow), Chronic conditions (up arrow), Uncompensated care (up arrow), Clinician cost (up arrow), and Rates and revenues (left-pointing arrow indicating compression). Source: HFS Research, 2026.

Margin deterioration

Demand for care will continue to rise as the population rapidly ages, bringing with it an increased prevalence of chronic conditions. Yet there will not be a commensurate increase in rates or revenues.

Increased uninsurance will triple the rate of uncompensated care, while the cost to attract and retain clinicians will grow only at rates above nominal inflation. Healthcare does not have a demand problem, just a monetization problem.

A directional table illustrating the margin deterioration threat. Five factors are listed with upward arrows indicating increases: Demand for care, Population age, Chronic conditions, Uncompensated care, and Clinician cost. A downward-pointing arrow leads to "Rates and revenues," shown with a left-pointing arrow to indicate that rates and revenues are not keeping pace with rising costs and demand. Source: HFS Research, 2026.

Clinician shortages

Burnout rates, slow speed to train replacements, and increasing demand for care are just some of the reasons clinician adequacy rates are likely to be in the high 80% range for physicians and in the low 90% range for registered nurses.

That rate will be worse for specialties most needed by the population based on demographics, such as primary care, geriatric care, and cardiology.

A two-part infographic illustrating the clinician shortage threat. On the left, three drivers are listed with upward arrows: Burnout rate, Training duration, and Demand. A downward-pointing arrow leads to the outcome panel on the right, which states current clinician adequacy rates: Physicians at high 80% and RNs at low 90%. Source: Urban Institute, Commonwealth Fund, HFS Research, 2026.

Patient expectations

Despite rising uninsurance rates, consumers remain primarily in control of who they see for care and when. In that context, their evolving expectations for speed, accuracy, and cost are critical to understand and address. The Kaiser Family Foundation reported that 36% of adults in the past 12 months have skipped or postponed care they needed because of cost. Other data indicate that up to 50% of consumers will switch doctors if their expectations are not met.

These factors can cost as much as 4% of annual revenue over 12 months and will compound if left unaddressed.

A two-part infographic illustrating the patient expectations threat. On the left, three factors are listed with upward arrows: Patient expectations, Delaying care, and Switching providers. A downward-pointing arrow leads to the outcome panel on the right, which states the revenue impact: a 4% decrease and compounding. Source: HFS Research, 2026.

Exhibit 1: Medicaid cuts will shave up to 20 pts of margins in 2026 alone

A grouped bar chart showing the financial impact of the elimination of Medicaid expansion across 40 states plus DC in 2026. Three metric pairs are displayed, each with a purple bar (With Medicaid) and an orange bar (Without Medicaid). Revenue: $1,128 with Medicaid versus $1,109 without Medicaid, a decrease of 1.7%. Cost: $1,086 with Medicaid versus $1,075 without Medicaid, a decrease of 1%. Margin: 3.75% with Medicaid versus 3.05% without Medicaid, a decrease of 18.6 percentage points. Revenue and cost figures are in millions. Source: Urban Institute, Commonwealth Fund, HFS Research, 2026.

Source: Urban Institute, Commonwealth Fund, HFS Research, 2026

Playbook benefits: Balancing the existential crisis of survival with the future of care delivery

Health system CIOs must lean into the business of care delivery. This playbook takes a programmatic approach, enabling CIOs to execute a strategic roadmap with contained solutions that optimize their existing assets while allowing AI to deliver significant business benefits in quick succession.

Benefit 1: Tangible and quantifiable business outcomes
  • Clear revenues lift through reduced denials and underpayments while accelerating cash realization through denial prevention and recovery, coding support, and contract variance detection.
  • Lower operating costs by reducing waste and unit costs via automation of the back office, supply chain anomaly detection, improving forecasting, and throughput improvements.
  • Higher workforce productivity by returning time to clinicians by cutting pajama time and admin burden through inbox message triage, ambient documentation, and grounded clinical summarization.
Benefit 2: Making AI count with legacy tech
  • Faster ROI by leveraging existing data flows with legacy systems already producing reliable signals such as 835/837, remits, and ADT feeds that AI can leverage without having to rip-and-replace.
  • Modular add-on delivery with tight integration, leveraging Services-as-Software as an AI-enabled delivery paradigm while avoiding getting stuck in platform re-architecture.
  • Maintaining operational control and auditability by developing governance loops so AI is trusted, safe, and scalable inside current operating models.
Benefit 3: Preparing for the future of care delivery
  • A command-center operating model to drive predictive orchestration of discharge barriers, LOS risk, and staffing and bed actions so that systems can manage increasing demand with constrained labor.
  • Digital-first patient with AI-driven access, intake, and front door capabilities to reduce leakage with no-shows, speed time-to-appointment, and improve self-serve completion.
  • More resilient care teams with scaled virtual nursing, AI monitoring, and closed-loop results follow-up to reclaim nurse time, reduce missed follow-ups, and raise patient trust as care shifts to a digital-biased hybrid.
The solution: AI-enabled modularity delivered by SaS

Care delivery CIOs must feel comfortable with an AI-enabled, modular approach to addressing the four key business challenges, rather than relying on legacy platform enablement. Modularity will not preclude tight integration; however, that integration must be considered notional by adopting Services-as-Software™ (SaS) as the delivery paradigm (see Exhibit 2).

Exhibit 2: Services-as-Software will evolve to become the default delivery paradigm to enable AI for the next 25 to 50 years

A transition diagram showing the evolution from traditional delivery to Services-as-Software™. On the left, a dashed-border box labeled "Traditional delivery" lists four elements with accompanying icons: People-based execution, Process outsourcing, and Static annual contracts. A large purple arrow in the center points right and is labeled "A new delivery paradigm inspired and enabled by AI." On the right, the Services-as-Software™ label anchors three new delivery characteristics: IP-led execution, Outcome orchestration, and Dynamic telemetry driven value realization, each shown with an icon. Source: HFS Research, 2026.

Source: HFS Research, 2026

The plan to execute practically

This playbook has identified 17 key solutions for delivering the largest bang for the buck in the least amount of time, leveraging existing tech-enabled capabilities and adopting SaS to deliver (see Exhibit 3). These solutions have been laid out in a practical sequence that will allow CIOs to achieve rapid ROI with lower risk, easier governance, and avoid the typical challenges of being stuck in pilot purgatory.

Exhibit 3: Delivering targeted business value with AI in a specific sequence to maximize outcomes

A sequenced roadmap matrix with 12 months on the horizontal axis and four key challenge rows on the vertical axis. The matrix is divided into four overlapping phases. Phase 1 "Fund the program" spans approximately months 1 to 3 and addresses Revenue headwinds with three solutions: 1. Denial prevention and recovery (predict, prioritize, and automate appeals); 2. AI-assisted CDI and coding specificity prompts; 3. Contract variance and underpayment detection. Phase 2 "Return time" spans approximately months 2 to 7 and addresses Clinician shortage with three solutions: 4. Ambient documentation; 5. In-basket message triage and draft responses; 6. Clinical summarization copilot. Phase 3 "Create capacity for margins" spans approximately months 4 to 10 and addresses Margin deterioration with five solutions: 7. Predictive capacity command center (flow, discharge barriers, bed and staff actions); 8. Perioperative optimization (OR utilization, cancellations); 9. Clinical variation analytics (cost variation and pathway nudges); 10. Supply chain forecasting and anomaly detection; 11. Back-office automation (AP/HR/IT ops). Phase 4 "Win the consumer" spans approximately months 7 to 12 and addresses Patient expectations with four solutions: 14. AI-driven access and leakage prevention (no-shows, referral routing); 15. AI digital front door (conversational access, intelligent routing, and enhanced self-service); 16. AI-assisted pre-registration and intake; 17. Cost transparency: estimate and explanation copilot. Solution 12 (Virtual nursing and AI-assisted monitoring) and Solution 13 (Closed-loop results follow-up) span across Phases 2 through 4. Source: HFS Research, 2026.

Source: HFS Research, 2026

The playbook: A roadmap that optimizes legacy with the potential of AI

This playbook has been developed with inputs from tens of practitioners, technologists, and healthcare experts. The intent is to rapidly deliver 17 solutions across four phases to directly address financial headwinds, margin deterioration, clinician shortage, and patient expectations.

Phase 1: Fund the program

Phase 1 addresses revenue headwinds. These four solutions fund the overall program and mitigate the need for capital expense or net new investments. You must implement three of these solutions in Phase 1; you can implement the fourth in Phase 4.

Solution 1: Denial prevention and recovery

A four-row structured solution card. Why this order matters: it delivers the fastest measurable financial ROI, has low patient safety risk, and uses mature RCM data flows. Do it right: integrate 835/837, remits, ERA/EOB, payer edits, and authorization data; create a denial taxonomy with root-cause mapping; deploy denial risk scoring and pre-bill work queues; automate appeal packet assembly with submission routing; develop a KPI loop including denial rate, overturn rate, and days to resolution. Expected outcomes: initial denial rate will decrease, denial overturn rate will increase, and days to resolution will decrease. Watch out for: bad taxonomy enabling incoherent denial reasons, payer policy drift where rules change and the model degrades, and gaming or cherry-picking easy denials to appeal with little net value. To-do list: Productionalized (live, in-production, and feeding work queues), Consistent results (three consecutive monthly cycles improved), Auditability (full audit trail for AI-generated packets). Source: HFS Research, 2026.

Solution 2: AI-assisted clinical documentation improvement (CDI) and coding specificity prompts

A four-row structured solution card. Why this order matters: it is dependent on workflow trust and governance that can be proven with denials; since CDI touches clinician documentation and compliance risk, executing denials first demonstrates operational control, auditability, and change management competence; it leverages benefits from the denial taxonomy and root-cause work. Do it right: standardize note templates and problem lists; enable structured-data capture including SNOMED/ICD mapping, orders, and HCC capture where relevant; deploy real-time documentation nudges in clinician workflow to facilitate CDI; include human-in-the-loop CDI validation with audit trail; measure CC/MCC capture rate, query rate, DNFB, and coding turnaround time. Expected outcomes: DNFB and coding lag will decrease, CC/MCC capture will increase, and rework and query loop time will decrease. Watch out for: compliance risk and upcoding perception, clinician fatigue from excessive prompts driving low adoption, and the risk of hallucinations or incorrect suggestions. To-do list: Core function (embedded in clinician and coder workflow), People matter (human-in-the-loop and QA operating), Compliance (compliance monitoring stable). Source: HFS Research, 2026.

Solution 3: Contract variance and underpayment detection

A four-row structured solution card. Why this order matters: underpayment requires accurate contract modeling, fee schedules, carve-outs, and clean payer configuration to avoid false positives and erode trust; it is easier to scale contract variances reliably with remit ingestion, taxonomy discipline, and queue operational stability. Do it right: build contract modeling that includes allowed amounts and DRG/APC logic; ingest remittances and contract terms into the analytics layer; deploy variance and anomaly detection rules; route to work queues and automate recovery letters; measure recovered amount per claim, variance backlog, and days to identify. Expected outcomes: recovery amounts per month will increase, identification of variances and anomalies will speed up, and variance backlog will decrease. Watch out for: false positives from incomplete contract modeling, missing data such as carve-outs, modifiers, and bundled payments, and slow cash realization if payer follow-up is weak. To-do list: Financial reconciliation (ensure expected vs. paid reconciliation works), Variance management (variances to create queue items with SLAs are in place), Cash (cash posted is tied to specific variance IDs). Source: HFS Research, 2026.

Phase 2: Return time

Given the shortage of clinicians, the idea is to optimize clinician time by returning time to them with these four solutions. We recommend executing the first three in Phase 2 and the fourth in Phase 3.

Solution 4: In-basket and message triage with draft responses

A four-row structured solution card. Why this order matters: it is before ambient documentation because it is simpler and faster to operationalize; inbox routing uses discrete message objects and clear telemetry and does not require audio capture or consent workflows; it can be a change-management wedge to build clinician trust in assistive AI with lower complexity. Do it right: define note standards and specialty templates; develop consent and recording governance; design tight EHR integration including note, orders, and coding support; support human verification, QA sampling, and guardrails; measure after-hours EHR, note closure time, and visits per session. Expected outcomes: handling time for responses will decrease, speed to first valid response will improve, and routing of clinicians will improve. Watch out for: privacy and consent concerns from patients or clinicians, note quality variability across specialties, and workflow friction reflected by extra clicks or slow turnaround. To-do list: Standards (routing and escalation path is approved), Safety (safety guardrails are validated), Metrics (measurement is live with patient experience). Source: HFS Research, 2026.

Solution 5: Ambient documentation

A four-row structured solution card. Why this order matters: ambient documentation requires consent, device logistics, specialty template tuning, and tighter compliance controls; running the inbox first creates readiness including governance and adoption habits, and reduces noise so ambient benefits are clearer. Do it right: standardize message categories and urgency rules; classify NLP with clear risk flags; route to team roles with draft responses; create safe-response policies with appropriate supervision and auditability; measure time to first response, handle time, and escalation rate. Expected outcomes: pajama time will decrease, note closure will speed up, and capacity for deployment to patient care will increase. Watch out for: safety risks such as missed urgent symptoms, wrong routing that can increase delays and frustration, and medical and legal concerns about AI-drafted messages. To-do list: Templates (EHR-integrated templates are operationalized), Predictability (stable QA and error taxonomy loop in place), Compliance (approved compliance and consent governance). Source: HFS Research, 2026.

Solution 6: Clinical summarization copilot with grounded chart synthesis

A four-row structured solution card. Why this order matters: it moves closer to clinical reasoning ahead of next phases; it enables grounded summarization pilots once governance and auditability are working. Do it right: enable creation of longitudinal patient records using CCD, FHIR, and internal data; record citations to source notes and results; secure with role-based access control (RBAC) and PHI controls and enable logging; deploy in workflow, e.g., SMART on FHIR visualized in EHR sidebar; measure time-to-decision, order latency, and satisfaction. Expected outcomes: clinicians spend less time finding key facts and duplicate work will decrease. Watch out for: hallucinations and inaccurate summaries, PHI exposure with access control gaps, and overreliance as clinicians stop checking sources. To-do list: Attribution (grounded citations to sources), Approvals (RBAC and PHI logging is signed off), Credibility (accuracy thresholds are sustained). Source: HFS Research, 2026.

Phase 3: Creating capacity for margins

Phase 3 addresses margin deterioration with seven solutions.

Solution 7: Predictive capacity command center (flow, discharge barriers, bed/staff actions)

A four-row structured solution card. Why this order matters: there is big margin impact, but the solution depends on higher integration and cross-functional workflow; it requires established and stable governance and integration proven in Phases 1 and 2. Do it right: enable real-time feeds including ADT, bed board, transport, EVS, lab and radiology, staffing, and case management; create an operational cockpit with a single pane of glass; enable AI for prediction, LOS and discharge risk, and boarding risk; embed playbooks across actions and activities; measure LOS, boarding hours, throughput, and agency spend. Expected outcomes: average length of stay will decrease, boarding time will decrease, and spending on premium labor will decrease. Watch out for: dashboards with insights but no actions, real-time feed fragility where ADT, EVS, and transport data latency breaks trust, and local optimization where one unit improves at the expense of another. To-do list: Standardization (running daily operations with playbooks), Actionable (80% of alerts have an action or closure), Improvements (KPI lift sustained over 8 to 12 weeks consistently). Source: HFS Research, 2026.

Solution 8: Perioperative optimization

A four-row structured solution card. Why this order matters: the operating room is typically considered the highest-value constrained asset and needs data normalization and governance. Do it right: normalize OR data such as usage duration, physician preferences, turnover, and staffing; develop predictive models for managing duration, cancellation risk, and utilization; deliver prescriptive recommendations on block release, staffing adjustment, and procedure sequencing; support enhanced OR governance with rules and scorecards; measure utilization, cancellation rate, and add-on capacity. Expected outcomes: usage of OR facilities and associated resources will increase, cancellations will decrease, and rate of on-time starts and on-schedule turnover will increase. Watch out for: surgeon resistance to block changes, bad duration predictions causing schedule chaos and distrust of AI, and incomplete perioperative data driving suboptimal decisions. To-do list: Schedules (prediction-model-driven scheduling for decisions is routine), Governance (complete governance is live and operational), Metrics (gains persist for up to 3 months consistently). Source: HFS Research, 2026.

Solution 9: Clinical variation analytics and pathway nudges

A four-row structured solution card. Why this order matters: it requires linking clinical and cost data with clinician governance; while a strong margin lever, it depends on data maturity and change management. Do it right: develop a data set linking clinical and cost data including costing, item-level supply, meds, labs, and imaging; enable AI detection of outliers of defined gold standard pathways and identify drivers; deploy nudges and review workflows; measure cost per case, utilization, and complications and readmissions. Expected outcomes: cost per case will decrease, unnecessary utilization will decrease, and quality will stabilize or improve. Watch out for: clinician distrust driven by algorithm-enabled healthcare cost-cutting, cost accounting mismatch due to wrong attribution driving inaccurate conclusions, and alert fatigue from too many nudges. To-do list: Gold standard (pathways to be used are approved and versioned), Metrics (adoption of embedded nudges is measured). Source: HFS Research, 2026.

Solution 10: Supply chain forecasting and anomaly detection

A four-row structured solution card. Why this order matters: material savings depend on item master hygiene and ERP integration; higher confidence returns when the organization is operating in a platform mode. Do it right: begin with item master cleanup including UDI in the ERP or supply chain system of record; integrate ERP, usage data, and physician preference cards; enable demand forecasting and set par levels to reduce wastage; detect and report price anomalies and substitutions and automate approvals; measure inventory turns, expirations, price variances (PPV), and stockouts. Expected outcomes: inventory waste will decrease, turns without stockouts will improve, and PPV anomalies will decrease. Watch out for: item master chaos due to duplicate SKUs and inconsistent units, clinician preference conflicts where substitution is rejected, and stockout risk if pars are cut too aggressively. To-do list: Technology (item master and ERP integration are stable), Analytics (forecasting drives pars used in practice), Metrics (savings validated with resolution workflows). Source: HFS Research, 2026.

Solution 11: Back-office automation

A four-row structured solution card. Why this order matters: there is low clinical risk, but enterprise workflow standardization is needed; it is easiest to scale after Phase 1 proves automation governance. Do it right: ingest all relevant documents such as invoices, contracts, and tickets; leverage AI extraction for classification and high confidence scoring; leverage industry-standard workflow routing with exceptions sent to humans and otherwise straight-through processing; embrace standard controls for audits and segregation of duties; measure the cost per invoice or ticket, cycle time, and hours saved. Expected outcomes: average cost of transactions across the supply chain landscape will decrease, cycle time to optimize inventory will decrease, and straight-through processing will increase. Watch out for: control failures that manifest in incorrect postings and audit findings, low accuracy due to significant exceptions that prevent standardization, and shadow workflows that bypass automation. To-do list: Operational (meets confidence thresholds with low exceptions in production), Compliance (separation of duties and compliance validated), Metrics (gains sustained across three purchase cycles). Source: HFS Research, 2026.

Solution 12: Virtual nursing and AI-assisted monitoring

A four-row structured solution card. Why this order matters: it requires operating model and device or network readiness with practical escalation pathways; it has a higher level of implementation complexity and requires an explicit business decision to adopt virtual nursing. Do it right: develop device and network readiness (QoS, Wi-Fi) to support virtual nursing and physician consultation; enable alert logic and escalation workflows; integrate tasks back into EHR; measure RN time reclaimed, falls, response times, and LOS. Expected outcomes: amount of nurse time reclaimed will increase, response to adverse events and identified alerts will be faster, and number of falls and adverse events will be reduced where applicable. Watch out for: operating model failure due to lack of clarity of roles and distrust, alarm fatigue due to too many alerts, and infrastructure issues with network or device uptime. To-do list: Op model (operating model is fully staffed), Stability (stable 99.999% uptime device, network, and EHR integration), KPI (sustained KPI and adoption thresholds met). Source: HFS Research, 2026.

Solution 13: Closed-loop results follow-up

A four-row structured solution card. Why this order matters: there must be established standardized rules for routing, outreach workflows, and audit dashboards given the high value for trust and safety. Do it right: normalize routing rules with critical value policies; set up AI-enabled triage for at-risk follow-ups; automate outreach by channel of choice, tasking, and escalations; create a dashboard for audit and compliance reporting; measure closed-loop percentage, time-to-follow-up, and missed follow-up. Expected outcomes: percentage of closed-loop interactions will increase, percentage of missed follow-ups will decrease, and speed to follow-up appointments will increase. Watch out for: false negatives such as missed abnormal results, lack of role clarity leading to accountability gaps, and outreach failures due to wrong contact info or lack of patient engagement. To-do list: Routing (approved standardized policies for routing), Engagement (automation of tasks for outreach), Auditability (safety monitoring with chart audits and approved fail-safes). Source: HFS Research, 2026.

Phase 4: Patient expectations

Patients’ evolving expectations are biased toward increased digital tool usage. We recommend these four solutions to address patients’ needs for faster care, higher diagnostic accuracy, and lower costs.

Solution 14: AI-driven access and leakage prevention

A four-row structured solution card. Why this order matters: it requires capacity management and RCM optimization; demand capture is improved after fixing throughput and staffing constraints; it depends on a clean provider directory and service catalog and integrated scheduling and referral data. Do it right: create a unified provider directory and services catalog; integrate scheduling, referral orders, and CRM for call center operations; enable predictive analytics for no-shows to allow strategic overbooking; incorporate referral routing intelligence; measure leakage percentage, fill rate, and no-show rate. Expected outcomes: patient no-show rate will decrease, loss of revenue due to missed appointments will decrease, and speed of time-to-appointment will increase. Watch out for: inequity where no-show models may penalize disadvantaged groups, operational mismatch where overbooking causes clinician burnout or longer waits, and directory and scheduling data quality issues. To-do list: Analytics (predictive analytics fully operationalized), Referrals (closed-loop referral tracking is consistent and stable), Equity (equity monitoring and an actionable override process have been established). Source: HFS Research, 2026.

Solution 15: AI-assisted pre-registration and intake

A four-row structured solution card. Why this order matters: patient-facing AI is best when internal operations can deliver on optimal capacity, scheduling integrity, and stable workflows. Do it right: digitally enable a standardized, version-controlled service catalog and reason-for-visit mapping; integrate scheduling with provider search and associated insurance rules; deploy conversational AI with containment analytics; allow for human handoff with context preserved; measure abandonment, self-serve completion, and time-to-schedule. Expected outcomes: registration accuracy will increase, time to check in will decrease, and volume of claim rework driven by registration errors will decrease. Watch out for: bad routing to the wrong specialty and location that could trigger safety issues, knowledge base drift such as when answers are not kept accurate and current, and suboptimal handoff to humans losing context and requiring rework. To-do list: Live (eligibility and document extraction is operational), Identity (patient identity and consent is validated). Source: HFS Research, 2026.

Solution 16: AI digital front door

A four-row structured solution card. Why this order matters: it relies on clean identity matching, eligibility integrations, and exception workflows; it is optimized with foundational controls and front-end standardization. Do it right: convert paper to digital and integrate patient registration forms with identity matching and eligibility; enable AI extraction of all patient information such as IDs, insurance, and payment type; develop real-time validation of coverage, demographics, and authorization triggers; measure reg accuracy, check-in time, auth delays, and claim rework. Expected outcomes: patient care abandonment rate will decrease, self-serve completion rate will increase, and time-to-schedule will improve. Watch out for: identity matching errors such as duplicate medical record numbers or wrong patient, eligibility inaccuracies leading to claims rejections, and limited patient digital savviness reducing completion. To-do list: Containment (validated end-to-end containment and handoff), Library (agreed to knowledge base and service catalog governance), Routing (confirmed safety routing). Source: HFS Research, 2026.

Solution 17: Price transparency

A four-row structured solution card. Why this order matters: it depends on reliable contract information, benefit logic, and robust billing operations. Do it right: modernize the patient portal with AI-enabled estimation to include covered benefits and out-of-pocket spend; enable digital, plain-language explanation of care and coverage with scenarios; include payment plans and financing workflows to drive further real-time self-service; integrate patient portal with call center CRM; measure estimate accuracy, pre-service collections, and complaints. Expected outcomes: estimate accuracy will improve, rate of pre-care collections and payment plan uptake will increase, and billing complaints will decrease. Watch out for: inaccurate estimates that will further erode trust, complex benefit designs that can create outliers, and regulatory and legal exposure due to misleading estimates. To-do list: Estimates (reconciles to posted claims within finance-approved tolerance), Compliance (estimate explanations are compliant and aligned with customer service scripts). Source: HFS Research, 2026.

Our perspective

The following list includes relevant HFS perspectives on care delivery operations in a shifting market; watch for our series on executing Services-as-Software within a health plan and on selecting the right partner for success.

Download Research

Author

Related


Download Research

    Sign In

    Sign up for a free
    research account

    With the exception of our Horizons reports, most of our research is available for free on our website. Sign up for a free account and start realizing the power of insights now.

    Digests/Newsletters: Overviews of the latest news, insight, and research by HFS.

    HFS Events: Exclusive invitations to HFS webinars, roundtables, and summits, bringing together key industry stakeholders focused on major innovations impacting business operations.

    By registering you agree to our privacy policy.

    I hereby consent that HFS Research can process my personal data.

    Premium Access

    Our premium subscription gives enterprise clients access to our complete library of proprietary research, direct access to our industry analysts, and other benefits.

    Contact us at [email protected] for more information on premium access.

    Help

    If you are looking for help getting in touch with someone from HFS, please click the chat button to the bottom right of your screen to start a conversation with a member of our team.

    [email protected]

      Contact Ask HFS AI Support