SAFE Note · 18-Month Runway · Terms available under NDA
Automating a costly, manual appeals process. ~$12B independent-lab revenue lost to denials annually (industry benchmark). Founded December 2025. First revenue target: under 8 months. Contingency model creates aligned incentives — labs pay only on recovery.
From raw EDI 835 denial file to recovered payment — an automated pipeline that replaces a costly manual process. Here's the full pipeline an investor should understand.
Lab uploads their ERA file (ANSI X12 835 standard). Ardia's parser extracts every Claim Adjustment Reason Code (CARC), denial amount, CPT code, and payer identifier, with a design target of under 0.3 seconds per file.
Ardia's deterministic rules engine cross-references each denied claim against 400+ Medicare policy Contractor (MAC) Local Coverage Determinations (LCDs). It flags whether the denial is compliant with CMS guidelines or incorrectly applied — the key to building a winnable appeal.
The Triadic Adjudicative Reasoning Architecture (TARA) combines large language model reasoning with symbolic logic rules. For each appealable claim, it: (1) retrieves relevant peer-reviewed evidence (NCCN, ASCO, AUA guidelines), (2) constructs a clinical narrative tailored to the denial code, (3) scores confidence 0–100%, and (4) drafts a CMS-formatted Level 1 appeal letter — targeting under 90 seconds end-to-end.
Texas SB 1188 (Sept 2025) and TRAIGA (Jan 2026) mandate that AI-generated healthcare decisions require human sign-off. Ardia's workflow surfaces the AI-drafted appeal to a licensed reviewer (billing specialist or lab director) with a one-click approve/edit/reject interface. This is designed to align Ardia with SB 1188's human-in-the-loop requirement from the start, rather than retrofitting it later.
Ardia submits the appeal, monitors ERA responses, auto-posts won payments, and collects its 15% contingency fee only when the lab recovers revenue. Every outcome — win, loss, partial — feeds back into the Payer Intelligence Engine, making future appeals smarter. This flywheel is the SaaS moat.
Six reasons this is the highest-conviction healthcare AI opportunity in the DFW market.
We are not aware of an AI-native platform focused specifically on independent-lab denial recovery. Waystar, R1, and AKASA all target hospitals. That gap is the opening Ardia is built to serve.
100% contingency model. Labs pay 15% only when they win. The no-upfront-cost model is designed to remove sales resistance — a lab risks nothing to try it. Whether that converts to signups is a hypothesis we plan to test in the DFW pilot.
TX SB 1188 + TRAIGA create compliance requirements competitors can't meet. PAMA 2027 forces labs to act. Ardia is the only compliant solution at the moment of crisis.
Every processed claim improves the Payer Intelligence Engine. By 100K claims, Ardia has the most comprehensive lab denial outcomes dataset in the US — unlocking a data licensing revenue stream worth 85–90% margin.
CAC of $3.5–8K, payback under 3 months, NRR target of 145%+ by Year 3. Rule of 40 score of 89 at scale (44% EBITDA + 45% growth). Best-in-class SaaS metrics for a health AI startup.
Architecture is payer-agnostic. Texas pilot → National 2027 → Canada/Mexico 2028 → Global 2030. The neuro-symbolic engine localizes to any MAC jurisdiction in weeks, not months.
Independent labs face an existential crisis: shrinking reimbursement, denial rates 2x industry average, and no viable appeal infrastructure.
Denial rates of 13.6–27% across independent labs, vs 11.8% national average (industry estimates). Labs are forced to absorb millions in unrecovered revenue.
Most labs lack the resources to file appeals. Yet 50–83% of appeals succeed when filed (industry estimates). Leaving $5–8B+ on the table annually.
Labs spend an estimated $19.7B annually overturning denials through manual labor. Appeals require 1–3 hours per claim. Unsustainable at scale.
15% annual Medicare lab cuts resume Jan 1 2027 (after 7 delays), escalating to 45% cumulative by 2029. Denials + cuts = lab closures.
Ardia sits at the intersection of three explosive growth curves — all underserved by current technology.
The market window is open, but closing. These four forces make this the highest-urgency health tech opportunity of 2026.
15% annual Medicare lab reimbursement cuts resume January 1, 2027, after 7-year delay. Escalates to 45% cumulative by 2029. The RESULTS Act remains unpassed. Labs must solve the denial crisis NOW or face shutdown.
Payer AI adoption grew 8 points to 34% full adoption in 2025. AI-generated denials require AI-powered appeals. Labs cannot compete with manual 1-3 hour appeals per claim. The denial game is now asymmetric without AI.
SB 1188 (Sept 2025) + TRAIGA (Jan 2026) mandate human-in-loop review and US data residency for healthcare AI. Ardia's architecture is designed to comply with these mandates from the start, which is harder for incumbents to retrofit. Texas labs gain a potential 12+ month advantage.
54% of digital health VC deployed in 2025. 83% premium per round for AI-enabled health startups. Healthcare AI funding hits all-time high. Capital for innovation is abundant. Deal multiples favor early movers with product-market fit.
| Feature | ✦ Ardia Precision Health | Waystar | R1 RCM | AKASA | Traditional Billing |
|---|---|---|---|---|---|
| Lab-Specific AI | Built for labs | Hospital | Hospital | Hospital | Rule-based |
| Molecular/Tox Coding | Native | Manual | |||
| LCD/NCD Rules Engine | 400+ policies | ||||
| TARA Reasoning | Explains decisions | Partial | |||
| TX SB 1188 Alignment | By design | TBD | TBD | TBD | |
| Success Fee Model | 15% recovered | License | License | License | % of billed |
| Independent Lab Focus | Only segment | Broad |
From contingency-fee appeals to platform SaaS to enterprise ACO contracts. Every tier eliminates customer risk and aligns Ardia's success with lab profitability.
Per-recovery model. No upfront cost.
Ideal for: Low-volume independent labs
Predictable monthly fees. Feature tiers.
Ideal for: Mid-market labs ($2–5M revenue)
Enterprise platform. Network analytics.
Ideal for: ACOs, Multi-lab networks
Ardia automates a costly manual appeals process. No upfront cost — labs pay a 15% contingency fee only when they recover revenue. Zero customer risk.
Monthly subscription by claim volume & feature tier. $3K–$15K/month.
15–20% of incremental recovered revenue. Zero customer risk.
De-identified claims + outcomes for pharma R&D and payer analytics.
Biomarker-to-trial matching revenue share with CROs and pharma sponsors.
Conservative unit economics. Payback under 3 months. Path to profitability by Year 3.
| Metric | 2026 | 2027 | 2028 | 2029 | 2030 |
|---|---|---|---|---|---|
| Paying Labs | 5 | 22 | 68 | 165 | 340 |
| ARR ($M) | $0.6 | $3.2 | $9.8 | $24.5 | $51.0 |
| Gross Margin | 62% | 70% | 75% | 78% | 80% |
| EBITDA Margin | (85%) | (22%) | 18% | 38% | 44% |
| Headcount | 4 | 12 | 28 | 55 | 95 |
RCM market shows 6–8× revenue multiples for AI-native platforms. Healthcare M&A remains at all-time highs.
AI-native health tech commands 83% premium per round in 2025. M&A multiples in RCM average 6–8× revenue. Strategic buyers (Optum, UnitedHealth, CVS) actively acquirer in healthcare AI space.
The planned allocation across the milestones that get Ardia to first revenue — engineering the demo engine and ToxIQ™ MVP, DFW pilot go-to-market, and the compliance and IP groundwork a healthcare-AI company needs.
Pricing model: a 15% contingency fee — paid only on recovered revenue. Illustrative unit economics: a mid-size lab (5,000–10,000 claims/mo, ~25% denial rate, ~65% of denials never appealed per MGMA 2023) has an estimated $812K–$1.6M in recoverable revenue per year; at 15% on a ~60% recovery that is roughly $73K–$146K per lab per year. This is an illustrative model, not a forecast of results — Ardia is pre-revenue.
Most decks hide the risks. Here are the five we take most seriously, each with the mitigation we're building against it.
We're raising $3–7M to take the demo engine and ToxIQ™ MVP to production, execute the DFW pilot, and reach first revenue — with 18 months of runway. Founded December 2025, Dallas-Fort Worth, Texas. Delaware C-Corporation. SAFE Note — terms available under NDA.
