You Can't Cut Your Way to Revenue Cycle Sustainability — Here's What Works Instead

Six revenue cycle leaders sat down at the HFMA September 2026 Annual Conference roundtable and arrived at the same conclusion: healthcare organizations cannot cut their way to sustainability. Not through headcount reduction. Not through outsourcing. Not through squeezing cost-to-collect below some arbitrary benchmark while revenue erodes underneath.

The data backs them up. Bad debt plus charity care climbed 14% year-over-year (Kaufman Hall / HFMA September 2026). Industry denial rates hit 9%, up 20% in three years (AMS Solutions 2026). Days in A/R stretched to 42. And $48 billion in net revenue was lost from denials and uncompensated care across 2,300+ hospitals — a 25% increase from the prior year. Cost-cutting does not fix revenue that was never captured in the first place.

What works instead: AI that protects and grows revenue — clean claims the first time, in-flight denial prevention, partnership accountability, and talent gap closure — combined with partnerships that are measured on revenue outcomes, not just cost metrics.

The Cost-Cutting Death Spiral

The instinct is logical: margins are compressed, so reduce costs. But in revenue cycle management, the cost-cutting playbook creates a compounding problem that accelerates the exact losses it was supposed to prevent.

Kevin Boren, East Region CFO at Essentia Health Duluth, framed it precisely at the HFMA roundtable:

"Cost is a huge challenge, margin compression is real, and everyone is measuring cost to collect. But you can't manage solely on cost to collect without suboptimizing revenue."

Here is how the spiral works:

• Cut billing staff → denial rates rise. Fewer billers means less pre-submission review, fewer appeals worked, and longer resolution cycles. The industry already sees 15% of claims initially denied (HFMA September 2026). Reducing the staff that prevent and resolve denials increases the denial rate — which increases the revenue loss that justified the cuts.
• Cut experienced billers → institutional knowledge disappears. With 30–40% annual billing staff turnover (HFMA 2026), practices lose payer-specific expertise faster than they can rebuild it. Every time a senior biller leaves, the replacement takes 6–12 months to learn the denial patterns, appeal strategies, and payer quirks that the departing employee handled instinctively.
• Cut follow-up resources → bad debt climbs. Bad debt plus charity care per calendar day rose 14% year-over-year in June 2026 (Kaufman Hall / HFMA). When organizations cut AR follow-up capacity, accounts age past timely filing deadlines. Revenue that could have been collected at 90 days becomes a write-off at 120.
• Cut technology investment → competitors pull ahead. Nearly half of healthcare executives identify revenue cycle as the top area for IT investment (HFMA September 2026). Organizations that defer AI deployment while cutting costs fall behind those deploying AI to capture revenue their manual processes cannot reach.

Adam Conley, leading a critical access hospital, stated what many small-market leaders already know:

"It's impossible for us to cut our way to sustainability. There aren't enough people and contracts to cut. We have to grow our way to sustainability."

The Revenue Side That Cost-Cutting Cannot Reach

The $48 billion in annual revenue losses does not come from overspending on billing operations. It comes from revenue that was earned but never collected — claims denied, payments delayed, appeals unworked, and coverage changes missed.

68% of revenue cycle leaders identified inaccurate or incomplete patient data at intake as the primary denial driver (Experian 2025 State of Claims). This is not a cost problem. It is a process problem that no amount of budget reduction fixes. The data must be right at intake, or the claim fails downstream.

Kenneth Hogue surfaced a dimension that cost-focused strategies miss entirely:

"You may lower claim denials but experience a rise in takebacks. You have to fight not only to capture revenue but also to keep it."

Payer takebacks — post-payment recoupment demands triggered by retrospective audits — represent a growing revenue threat that operates outside the traditional denial management framework. A practice can achieve a 5% denial rate and still lose six figures to takebacks from claims that were paid, then clawed back months later. AI that monitors payer behavior, identifies takeback patterns, and builds documentation defenses proactively addresses a revenue protection layer that manual teams and cost-cutting strategies both ignore.

AI Replaces the Work That Cannot Be Staffed

The talent pipeline problem has passed the point where hiring — at any cost — solves it. Desmond Jackson, leading a rural hospital's revenue cycle, described the structural constraint:

"Can we leverage AI and technology in a way that does not increase hiring costs? We're in kind of a labor desert and don't have a pipeline of revenue cycle talent."

Jeff Costello reinforced the same reality from a different market:

"In our market, we can't hire problems away. There's just too much need and not enough revenue cycle talent available. We are open to how we close gaps — whether locally, overseas, or with the help of AI."

The talent gap is not a temporary recruitment challenge. It is a structural market condition driven by competition from every industry that needs analytical workers, combined with the specialized knowledge required for healthcare billing. When a medical billing specialist can earn comparable wages in a less complex industry, the pipeline empties permanently.

AI Versus Offshoring: The Target Has Shifted

Dennis Jones provided the clearest assessment of where AI disrupts the labor model:

"AI-driven layoffs are going to affect offshoring companies more… At 7 p.m., you see thousands of people pouring into buildings to do repetitive, high-volume, sometimes low-balance work. That's the AI target."

The offshore model was healthcare's answer to domestic labor shortages: move repetitive claims work to markets with available labor at lower cost. AI changes the calculus by eliminating the repetitive work entirely — not relocating it. Prior authorization submission, insurance verification, claim scrubbing, payment posting, and low-balance AR follow-up are all categories where AI achieves higher accuracy than manual processing with zero incremental labor cost per transaction.

This does not mean eliminating billing teams. It means redirecting them. The HFMA roundtable consensus was clear: AI handles the volume while skilled staff handle the exceptions. The practices that try to handle both volume and exceptions with the same constrained team lose revenue on both fronts.

Clean Claims the First Time: The AI Revenue Protection Layer

The single highest-impact AI capability for revenue cycle sustainability is making claims clean before they are submitted. Hospitals currently spend $19.7 billion per year on appeals — and only half are overturned (HFMA September 2026). Every dollar spent on appeals is a dollar that could have been saved by preventing the denial.

Dennis Jones described the shift happening in real time:

"Looking at AI to make the claims process clean the first time."

Clean-first AI operates across three layers that manual processes cannot sustain simultaneously:

1. Upstream prevention. AI verifies eligibility, confirms coverage, checks prior authorization requirements, and validates patient data at scheduling and registration — catching the intake errors that cause 68% of denials before the patient is even seen.
2. In-flight intervention. During the billing workflow — documentation, coding, charge capture, claim assembly — AI applies real-time payer intelligence to flag denial risks inside active workflows. The shift from "retrospective visibility to in-flight intervention" is the architectural change that HFMA identified as the 2026 priority.
3. Post-submission defense. For claims that do encounter payer resistance, AI generates appeals within hours instead of weeks, works 100% of denied claims instead of the 35% that manual teams reach, and tracks payer behavior patterns to prevent recurring denial categories from reappearing.

The OIG's September 2026 federal review underscores why prevention matters more than recovery: 97% of appealed Medicare Advantage SNF authorization denials from one major contractor were ultimately overturned — but only 18% of denials were ever appealed. The revenue was recoverable. The organizations simply did not have the staff bandwidth to pursue it. AI that prevents denials upstream eliminates the need for appeals entirely. AI that automates appeals for the denials that do occur ensures 100% of recoverable revenue is pursued.

Partnership Accountability: Trust but Verify

Revenue cycle sustainability is not purely a technology problem. It is also a partnership accountability problem. Kenneth Hogue articulated the standard that AI enables:

"If someone on the vendor's team has moved on, I would rather the vendor tell me. What I don't want is no communication."

And more directly: "If the health system has to surface the problem, that's not client management."

Traditional billing company relationships operate on trust: the practice sends claims, the billing company processes them, and the practice receives reports. The problem is that those reports are self-generated by the entity being measured. When denial rates rise or AR days extend, the billing company controls the narrative.

AI-driven accountability changes this dynamic by providing the health system with independent, real-time visibility into every claim, every denial, every appeal, and every payment — regardless of who processes the work. The AI does not depend on the vendor's reporting accuracy. It generates its own data from the same source systems, creating an independent verification layer that makes partnership accountability automatic rather than adversarial.

This is the billing company replacement conversation happening across healthcare in 2026 — not replacing billing companies with AI, but replacing trust-based relationships with AI-verified accountability that makes every partner's performance measurable in real time.

The Build-vs-Buy Debate: Waiting 18 Months Has a Revenue Cost

Dennis Jones raised the question every health system evaluating AI must answer:

"Epic might take 18 months or longer. All these smaller companies are knocking on our door with solutions right now. But there's value in waiting for stem-to-stern platform."

The trade-off is real. Platform integration has long-term advantages: unified data, consistent workflows, reduced vendor management overhead. But the revenue losses are compounding now — $48 billion annually and growing 25% year-over-year. An 18-month wait at current loss rates represents billions in aggregate revenue that could have been protected by deploying AI solutions today.

The practical answer for most organizations is not either/or. Deploy AI that integrates with current workflows now. Capture the revenue that manual processes are leaving on the table. Maintain architectural flexibility to adopt platform-native capabilities as they mature. The organizations that wait for the perfect platform while losing 9% of claims to denials and watching bad debt climb 14% annually are optimizing for future architecture at the cost of present revenue.

Bad Debt Is the Quiet Emergency

While denials get the headlines, bad debt is the accelerating crisis that compounds every other revenue challenge. The Kaufman Hall / HFMA September 2026 data paints an urgent picture:

• Bad debt + charity care per calendar day up 14% year-over-year
• Bad debt as a percentage of gross revenue rose 3%
• Finance leaders now hope to collect 5–10 cents on the dollar for uninsured patients (Todd Nelson, HFMA)
• H.R.1 / One Big Beautiful Bill Act Medicaid funding cuts expected to accelerate the payer mix erosion into 2027

Erik Swanson of Kaufman Hall warned that the "biggest impacts" have not yet materialized. As patients shift from commercial coverage to governmental plans and the uninsured population grows following ACA subsidy expiration, the payer mix deterioration creates a structural revenue decline that no amount of billing efficiency can offset.

AI insurance discovery and financial clearance address the bad debt problem at the point of service — identifying patients with unknown coverage, discovering secondary insurance, estimating patient responsibility accurately, and securing payment arrangements before services are rendered. For organizations collecting 5–10 cents on the dollar post-service, moving financial clearance to pre-service represents a 5–10x improvement in collection rates on the same patient population.

The Sustainability Model That Actually Works

The HFMA roundtable consensus points to a revenue cycle sustainability model built on three pillars:

Matt Leshy, Provider Practice Leader at Signature Performance, summarized it at the roundtable's opening:

"Health systems cannot cut their way to sustainability. They must protect and grow revenue while using technology responsibly, addressing workforce constraints and creating more accountable partnerships."

The organizations that treat AI as another cost-reduction tool miss the point. AI is the revenue protection layer that makes the entire revenue cycle sustainable — not by making billing cheaper, but by making revenue capture more complete, more consistent, and more defensible against the payer strategies designed to erode it.

The math is not complicated. $48 billion in lost revenue. 15% initial denial rate. 14% bad debt growth. 30–40% staff turnover. Cost-cutting does not solve any of these. AI-driven revenue protection, clean-first claims processing, and accountable partnerships solve all of them.

See how BAM AI builds the revenue protection layer — from upstream denial prevention to in-flight intervention to partnership accountability.