A national healthcare organization providing end-to-end billing and accounts receivable services for prominent hospitals and large physician groups — processing high volumes of paper EOBs and payer correspondence across a complex, multi-payer environment.
The client’s operations leadership knew they had a denial recovery problem — but assumed it was a staffing issue. An internal audit revealed something more specific: over 10% of total denials from paper remittances were never being identified in the first place. They weren’t being ignored — they were invisible. Because paper EOBs were posted manually, denial information embedded in those remittances was never systematically captured or routed. The denials simply didn’t exist in the workflow.
The downstream consequences were compounding. Because manual posting was slow, secondary claims and patient billing couldn’t be initiated until primary payments cleared — creating a cash flow bottleneck that grew worse during high-volume periods. Meanwhile, payer correspondence was being sorted inconsistently, and missed response deadlines were generating avoidable rework.
The client had tried adding headcount, but manual posting carried high turnover. Every new hire required weeks of training before reaching acceptable accuracy — and by the time they did, a significant percentage had already left.
NDS was brought in specifically to solve the paper EOB problem — not to overhaul the client’s entire posting operation. The initial scope was narrow: convert paper EOBs into accurate, postable 835 ERA files so that denials from paper remittances would be captured and routed like any electronic remittance.
The AI used supervised deep learning trained on NDS’s proprietary payer remittance datasets to read, extract, and validate payment data from paper EOBs — including deciphering complex payer-specific denial codes and mapping them to standardized ANSI codes. Intelligent claim cross-referencing filled in missing data elements from the original claim to ensure each 835 file was complete.
What the client hadn’t anticipated was the secondary effect. Once posting was automated, the manual team bottleneck disappeared — and with it, the delays to secondary billing and patient billing. Cash flow improved not because of a new collections initiative, but because payments were simply being processed faster.
Midway through implementation, NDS added automated correspondence indexing — classifying incoming payer documents by type, extracting key data fields, and hosting them on a searchable web-based Document Management System. This wasn’t in the original scope. The client’s appeals team requested it after realizing that the newly surfaced paper EOB denials needed supporting correspondence to be worked — and that correspondence was scattered across physical files and shared drives.
Pilot on a subset of payers to validate extraction accuracy. Initial accuracy on one regional payer was lower than expected due to a non-standard EOB format — NDS built custom extraction rules that resolved it within 10 days.
Expanded to full payer mix. Correspondence indexing added in parallel at the client’s request.
Full production with ongoing reinforcement learning from posting outcomes.
Improvement in denial recovery — all paper EOB denials now captured and entering the workflow for the first time
Reduction in payment posting costs — eliminated the recurring cycle of hiring, training, and losing manual posting staff
Improvement in denial appeal turnaround — driven by the correspondence indexing the client hadn’t originally requested
Significant improvement from faster secondary billing — the CFO attributed it to “removing a bottleneck we’d been working around for years”
Unexpected outcome: The AI surfaced a pattern of modifier misapplication on a specific payer that had gone undetected for over a year. The client's billing team used this to recover underpayments on previously adjudicated claims — a recovery effort that wasn't part of the original project scope.
What the Client Said
"We brought NDS in to fix a paper EOB problem. What we didn’t expect was how much of our cash flow delay was tied to the same bottleneck. The denial recovery was the headline win, but the operational cleanup underneath it was just as valuable."
Discover the way forward through more case studies.
Share your sample data with us and we’ll build a proof of concept showing exactly how NDS handles your EOBs and ERAs — from extraction to a clean, postable 835 — tailored to your payer mix and PM system.
© 2026 NDS InfoServ. All rights reserved.