Signal/capture-signal-cgm-assets.sh

#!/usr/bin/env bash
# =============================================================================
# capture-signal-cgm-assets.sh
# Captures Signal CGM strategic analysis outputs, commits to Git, syncs Obsidian
# Idempotent — safe to run multiple times
# =============================================================================
set -euo pipefail

# ── Configuration ─────────────────────────────────────────────────────────────
SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
REPO_ROOT="$SCRIPT_DIR"
OUTPUT_DIR="$REPO_ROOT/CGM-Denial-Prevention/01-Claude-Outputs"
ANALYSIS_DIR="$OUTPUT_DIR/Analysis"
ASSETS_DIR="$OUTPUT_DIR/Assets"
OBSIDIAN_DIR="/Users/sttil-solutions/Documents/Obsidian_Vault/STTIL-Vault/Projects/DMEPOS/CGM-Analysis"

# ── Helpers ───────────────────────────────────────────────────────────────────
log()  { echo "  → $*"; }
ok()   { echo "  ✓ $*"; }
fail() { echo "  ✗ $*" >&2; exit 1; }

echo ""
echo "━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━"
echo "  Signal CGM — Asset Capture Script"
echo "━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━"

# ── Step 1: Directory Structure ───────────────────────────────────────────────
log "Creating directory structure..."
mkdir -p "$ANALYSIS_DIR" "$ASSETS_DIR"
ok "Directories ready"

# =============================================================================
# FILE 61 — signal-cgm-segment-scoring-v1.md
# Source: "Re-rank the customer segments using an explicit scoring model"
# =============================================================================
log "Writing signal-cgm-segment-scoring-v1.md..."
cat > "$ANALYSIS_DIR/signal-cgm-segment-scoring-v1.md" << 'ENDDOC'
# Signal CGM — Segment Scoring Model v1
### Explicit Criteria, Weights, and Rankings Across Three Objectives
### STTIL Solutions LLC | April 2026

---

## Scoring Architecture

Three independent rankings. Each uses different criteria, weights, and success
definitions. A segment that ranks #1 for SaaS ICP may be wrong for a pilot and
irrelevant for an asset sale. Treat them as separate decisions.

---

## Model 1: Best SaaS ICP for MRR/ARR

| Criterion | Weight | What it measures |
|-----------|--------|-----------------|
| Pain intensity | 25% | How acutely the segment feels the CGM denial problem |
| Willingness / ability to pay | 20% | Named budget, ROI clarity, buyer who can sign |
| Revenue per customer (ARR × LTV) | 20% | Monthly contract × expected contract duration |
| Sales cycle speed | 15% | Weeks from first contact to signed contract |
| Churn durability | 10% | Still exists and values the tool in 24 months? |
| Reachability at scale | 10% | Find and reach efficiently without large sales org |

### SaaS ICP Scores

| Segment | Pain (25%) | Pay (20%) | ARR×LTV (20%) | Cycle (15%) | Churn (10%) | Reach (10%) | Score |
|---------|-----------|-----------|--------------|-------------|------------|-------------|-------|
| Mid-size supplier (10–50 emp) | 9 | 8 | 8 | 7 | 6 | 7 | **7.80** |
| Billing company / DME RCM | 8 | 7 | 7 | 5 | 8 | 6 | **6.95** |
| Small supplier (2–8 emp) | 10 | 5 | 4 | 6 | 4 | 8 | **6.40** |
| VGM Group / MSO | 5 | 4 | 7 | 3 | 9 | 5 | **5.30** |
| DME platform vendor (NikoHealth) | 3 | 4 | 8 | 2 | 9 | 3 | **4.65** |
| State / national association | 2 | 1 | 1 | 3 | 7 | 9 | **2.55** |
| Grant-funded / QI org | 3 | 2 | 2 | 2 | 4 | 4 | **2.60** |

**Key rationale:**
- Mid-size supplier leads because they have pain + budget + ROI clarity at $199–$399/month
- Small supplier has maximum pain (10/10) but minimum reliability — high churn risk as CB 2028 approaches
- Billing company scores #2: absorbs denial labor directly; LTV longer than any individual supplier
- NikoHealth scores low on SaaS — they are an asset buyer, not a subscriber

---

## Model 2: Best Buyer for Asset Sale / Strategic Handoff

| Criterion | Weight | What it measures |
|-----------|--------|-----------------|
| Distribution reach | 30% | How many suppliers does this buyer already reach? |
| Strategic need | 25% | Documented feature gap; would they build it otherwise? |
| Price ceiling | 20% | How much would a motivated buyer plausibly pay? |
| Speed to close | 15% | Weeks from conversation to signed term sheet |
| Build vs. buy calculus | 10% | Is acquiring faster than building given CB 2028 window? |

### Asset Sale Scores

| Segment | Distribution (30%) | Need (25%) | Price (20%) | Speed (15%) | B/B (10%) | Score |
|---------|-------------------|-----------|------------|-------------|-----------|-------|
| DME platform vendor (NikoHealth) | 10 | 9 | 9 | 3 | 7 | **8.20** |
| VGM Group / MSO | 10 | 7 | 7 | 4 | 8 | **7.55** |
| Billing company / DME RCM | 7 | 8 | 6 | 5 | 7 | **6.75** |
| State / national association | 8 | 4 | 2 | 2 | 3 | **4.40** |
| Mid-size supplier | 1 | 5 | 2 | 4 | 3 | **2.85** |
| Small supplier | 1 | 4 | 1 | 3 | 2 | **2.10** |
| Grant-funded / QI org | 3 | 3 | 2 | 1 | 2 | **2.40** |

**NikoHealth (#1 asset buyer) rationale:**
- Already serves the exact buyer profile; CGM feature gap is documented and real
- API-first architecture makes integration technically trivial
- CB 2028 window makes buying faster than 6–9 month internal build
- Price ceiling is 3–5× current $25K–$60K ask for a motivated platform buyer

---

## Model 3: Best Pilot Validation Partner

| Criterion | Weight | What it measures |
|-----------|--------|-----------------|
| Data accessibility | 25% | Structured CGM billing data with denial reason codes |
| Cooperation likelihood | 20% | Will they actively participate and give feedback? |
| Signal quality | 20% | Will pilot results generalize to target market? |
| PHI / compliance overhead | 20% | BAA and data security burden |
| Feedback loop speed | 15% | How quickly is denial rate change measurable? |

### Pilot Scores

| Segment | Data (25%) | Coop (20%) | Signal (20%) | PHI (20%) | Speed (15%) | Score |
|---------|-----------|-----------|-------------|----------|------------|-------|
| Billing company / DME RCM | 10 | 7 | 10 | 6 | 6 | **8.00** |
| Mid-size supplier | 8 | 8 | 9 | 5 | 8 | **7.60** |
| DME platform vendor | 10 | 4 | 10 | 7 | 4 | **7.30** |
| Small supplier | 6 | 9 | 7 | 5 | 7 | **6.75** |
| VGM / MSO | 3 | 5 | 4 | 4 | 3 | **3.80** |

**Billing company leads pilot scoring** because they have multi-supplier,
multi-plan, multi-jurisdiction billing data — the fastest path to H1 validation
(denials are documentation-fixable) across a meaningful sample size.

---

## Government Enforcement Context

| Finding | Source | Strategic Implication |
|---------|--------|----------------------|
| CGM improper payment rate: 25.2% / $278.5M/yr | CMS 2024 | One in four CGM dollars improperly paid |
| 94.2% from documentation failures | CMS 2024 | Not fraud — fixable workflow gaps |
| $1.9B total DMEPOS improper payments FY2024 | OIG | CGM is highest-scrutiny category |
| $1.8B in payments suspended 2025 | CMS Fraud Defense Ops | Enforcement is executing at scale now |
| CGM as explicit 2026 enforcement priority | OIG / DOJ | Legitimate suppliers get caught in sweeps |
| Unequal enforcement: pharmacy vs DMEPOS | CMS-2025-0242-0025 | DMEPOS held to higher standard than pharmacy for same product |

**The dual-edge positioning:** Signal CGM's audit log is not just a billing tool —
it is liability documentation. A supplier who can show time-stamped pre-submission
checks has a defensibility argument when the MAC issues an ADR.

---

## NikoHealth Foothold Assessment

**Verdict: Credible and growing challenger. Not the market standard.**

| Indicator | Assessment |
|-----------|-----------|
| Architecture | Cloud-native, API-first — genuine advantage over Brightree |
| Pricing | More accessible for small suppliers than Brightree ($600–$1,500+/mo) |
| Market position | Capturing switchers from legacy platforms; not dominant |
| Customer count | G2 review volume suggests low hundreds, not thousands |
| CGM-specific intelligence | Generic authorization alerts; no 6-month visit tracking, no MAC-jurisdiction rules, no 45-day runway logic |
| Competitive risk | Could close the CGM gap in 6–9 months of focused engineering |

**Asset sale timing window: open but not permanent.** NikoHealth's API-first
architecture makes acquisition or licensing integration technically trivial
relative to Brightree's legacy stack.

---

## Sources
- [CMS 2024 CGM Improper Payment Data](https://www.cms.gov/training-education/medicare-learning-networkr-mln/compliance/medicare-provider-compliance-tips/glucose-monitoring-supplies)
- [OIG 2025: CGM Payments Exceeded Supplier Costs](https://oig.hhs.gov/reports/all/2025/medicare-payments-for-continuous-glucose-monitors-and-supplies-exceeded-supplier-costs-and-retail-market-prices-indicating-medicare-can-save-at-least-tens-of-millions-of-dollars-in-one-year/)
- [KFF: MA Prior Authorization 2024](https://www.kff.org/medicare/nearly-50-million-prior-authorization-requests-were-sent-to-medicare-advantage-insurers-in-2024/)
- [Unequal DME/Pharmacy Enforcement — CMS Comment Record](https://downloads.regulations.gov/CMS-2025-0242-0025/attachment_1.pdf)
- [NikoHealth G2 Reviews 2026](https://www.g2.com/products/nikohealth/reviews)
- [DME/HME Software 2026 — Coruzant](https://coruzant.com/software/dme-and-hme-software-in-2026/)
- [DOJ $14.6B Healthcare Fraud Takedown 2025](https://www.justice.gov/opa/pr/national-health-care-fraud-takedown-results-324-defendants-charged-connection-over-146-billion)
- [Federal Authorities Targeting CGM Reimbursement — Nat'l Law Review](https://natlawreview.com/article/federal-authorities-are-targeting-continuous-glucose-monitoring-cgm-device)
ENDDOC
ok "signal-cgm-segment-scoring-v1.md written"

# =============================================================================
# FILE 62 — signal-cgm-re-scored-composite-v2.md
# Source: "Take the prior scoring model and re-run it with new priority order"
# =============================================================================
log "Writing signal-cgm-re-scored-composite-v2.md..."
cat > "$ANALYSIS_DIR/signal-cgm-re-scored-composite-v2.md" << 'ENDDOC'
# Signal CGM — Re-Scored Composite Model v2
### Asset Sale Primary (50%) · Pilot Secondary (30%) · SaaS Tertiary (20%)
### MA + Medicaid First Proof-of-Concept Payer Scope
### STTIL Solutions LLC | April 2026

---

## Priority Weights

| Objective | Weight | Rationale |
|-----------|--------|-----------|
| Asset sale / strategic handoff | 50% | Fastest path to capital realization |
| Pilot validation | 30% | Evidence from pilot raises asset sale price |
| SaaS ICP potential | 20% | Informs asset buyer's deployment ROI |

---

## Scoring Assumption Audit

### Asset Sale — Why These Weights

**Distribution reach (30%):** An asset sale to a platform touching 500+ suppliers
is worth orders of magnitude more than one touching one. This is the dominant
criterion because it determines the multiplier on the asset's downstream value.

**Strategic need (25%):** Pull motion (buyer has documented gap) is more reliable
than push motion (we convince buyer they need it). NikoHealth's CGM intelligence
gap is documented and real. Billing companies' MA plan-rule complexity is felt daily.

**Price ceiling (20%):** Current $25K–$60K is priced for a direct supplier buyer.
A platform vendor or MSO can justify 3–5× that. Weight at 20% because price ceiling
is contingent on strategic need — no need, no premium.

**Speed to close (15%):** Closing faster is better but not at the expense of deal
size. Pursue fast-close candidates in parallel with long-cycle targets.

**Build vs. buy (10%):** Modifier on other criteria. Accelerates a deal that
already has strategic need; doesn't create one where those are absent.

### Pilot — Why These Weights

**Data accessibility (25%):** Pilot must generate measurable denial rate change.
Requires structured CGM billing data with denial reason codes — not spreadsheets.

**Cooperation likelihood (20%):** Active partner engagement multiplies data access
value. Poor data + high cooperation still generates useful qualitative signal.

**Signal quality (20%):** Pilot must generalize to the asset buyer's customer base.
Single Florida supplier = anecdote. Billing company across 3 MACs = evidence.

**PHI/compliance overhead (20%):** High-overhead pilot = delayed pilot. Given CB
2028 window, compliance friction is a real timeline risk.

**Feedback loop speed (15%):** A 30–60 day feedback cycle is acceptable;
6 months is not viable given the urgency window.

---

## Individual Segment Scores — MA + Medicaid Scope Applied

### NikoHealth-Type Platform Vendor

**Asset Sale Sub-Score:**
| Criterion | Wt | Score | Note |
|-----------|-----|-------|------|
| Distribution reach | 30% | 10 | Serves exact buyer profile across hundreds of suppliers |
| Strategic need | 25% | 10 (+1) | MA+Medicaid scope widens the gap — no platform has plan-specific rules |
| Price ceiling | 20% | 9 | MA module = 3–5× current ask to a motivated platform buyer |
| Speed to close | 15% | 3 | Enterprise procurement: 4–8 months minimum |
| Build vs. buy | 10% | 8 (+1) | MA plan-rule maintenance is ongoing data work, not a sprint |
**Sub-score: 8.55**

**Pilot Sub-Score:**
| Criterion | Wt | Score | Note |
|-----------|-----|-------|------|
| Data accessibility | 25% | 10 | Best dataset: all supplier clients across all payers |
| Cooperation likelihood | 20% | 3 (−1) | Exposing own platform gap is sensitive under MA scope |
| Signal quality | 20% | 10 | Multi-supplier, multi-plan, multi-jurisdiction |
| PHI overhead | 20% | 7 | HIPAA infrastructure exists |
| Feedback speed | 15% | 4 | Enterprise review before data access |
**Sub-score: 7.10**

**SaaS ICP Score: 4.65**
**NikoHealth Composite: (8.55×0.50) + (7.10×0.30) + (4.65×0.20) = 7.34**

---

### Billing Company / DME RCM Outsourcer

**Asset Sale Sub-Score:**
| Criterion | Wt | Score | Note |
|-----------|-----|-------|------|
| Distribution reach | 30% | 7 | Serves 10–50 DMEPOS clients per firm |
| Strategic need | 25% | 9 (+1) | MA plan-specific PA rules are their most expensive unsolved problem |
| Price ceiling | 20% | 7 (+1) | $40K–$80K justified across 20+ clients under MA scope |
| Speed to close | 15% | 5 | Owner/managing partner: 60–90 day decision cycle |
| Build vs. buy | 10% | 7 | Not a tech company; will buy |
**Sub-score: 7.20**

**Pilot Sub-Score:**
| Criterion | Wt | Score | Note |
|-----------|-----|-------|------|
| Data accessibility | 25% | 10 | MA + Medicaid billing data across all clients — defines the advantage |
| Cooperation likelihood | 20% | 8 (+1) | MA complexity is their daily pain; highly motivated under this payer scope |
| Signal quality | 20% | 10 | Multi-supplier, multi-plan data |
| PHI overhead | 20% | 6 | HIPAA infrastructure exists; BAA manageable |
| Feedback speed | 15% | 6 | Fast initiation; first MA cycle data in 30 days |
**Sub-score: 8.20**

**SaaS ICP Score: 7.75 (+0.80 under MA scope — tool becomes a revenue line)**
**Billing Company Composite: (7.20×0.50) + (8.20×0.30) + (7.75×0.20) = 7.61**

---

### VGM Group / MSO

**Asset Sale Sub-Score:**
| Criterion | Wt | Score | Note |
|-----------|-----|-------|------|
| Distribution reach | 30% | 10 | Thousands of small-to-mid DMEPOS suppliers nationally |
| Strategic need | 25% | 6 (−1) | MA/Medicaid are plan-specific problems outside VGM's policy leverage |
| Price ceiling | 20% | 7 | $75K–$150K justified as member benefit |
| Speed to close | 15% | 3 | 6–12 month committee procurement |
| Build vs. buy | 10% | 8 | VGM endorses/buys tools, does not build |
**Sub-score: 7.15**

**Pilot Sub-Score:**
| Criterion | Wt | Score | Note |
|-----------|-----|-------|------|
| Data accessibility | 25% | 3 | No supplier billing data; must recruit members |
| Cooperation likelihood | 20% | 5 | Conceptual support; cannot drive participation |
| Signal quality | 20% | 3 (−1) | Member base skews rural/small; less MA density |
| PHI overhead | 20% | 4 | VGM as intermediary adds complexity layer |
| Feedback speed | 15% | 3 | Slow: member recruitment + separate BAAs |
**Sub-score: 3.60**

**SaaS ICP Score: 5.30**
**VGM Composite: (7.15×0.50) + (3.60×0.30) + (5.30×0.20) = 5.72**

---

### Mid-Size Independent Supplier

**Asset Sale Sub-Score:**
| Criterion | Wt | Score | Note |
|-----------|-----|-------|------|
| Distribution reach | 30% | 1 | Single supplier; no scale |
| Strategic need | 25% | 6 (+1) | MA scope raises personal pain — daily multi-plan complexity |
| Price ceiling | 20% | 3 (+1) | MA ROI strengthens argument; still marginal for asset sale |
| Speed to close | 15% | 4 | Owner decision: 2–8 weeks if motivated |
| Build vs. buy | 10% | 3 | Cannot build; would subscribe not buy |
**Sub-score: 3.30**

**Pilot Sub-Score:**
| Criterion | Wt | Score | Note |
|-----------|-----|-------|------|
| Data accessibility | 25% | 7 | Structured in Brightree/NikoHealth; needs 50+ MA patients for signal |
| Cooperation likelihood | 20% | 8 | Billing manager engages daily; highly motivated |
| Signal quality | 20% | 7 (−2) | Thin MA mix reduces cross-plan generalizability |
| PHI overhead | 20% | 5 | BAA: standard, manageable |
| Feedback speed | 15% | 8 | Monthly MA billing cycle; fast iteration |
**Sub-score: 6.95**

**SaaS ICP Score: 8.10 (+0.30 — MA complexity is their most expensive unsolved problem)**
**Mid-Size Supplier Composite: (3.30×0.50) + (6.95×0.30) + (8.10×0.20) = 5.36**

---

## Composite Ranking Summary (Base + MA/Medicaid Adjusted)

| Rank | Segment | Base Composite | MA+Med Adj | Final Score |
|------|---------|---------------|-----------|-------------|
| #1 | Billing company / DME RCM | 7.17 | +0.44 | **7.61** |
| #2 | NikoHealth-type platform | 7.22 | +0.12 | **7.34** |
| #3 | VGM Group / MSO | 5.98 | −0.26 | **5.72** |
| #4 | Mid-size independent supplier | 5.27 | +0.09 | **5.36** |

---

## The Sequencing Logic

The #1 and #2 positions are 0.27 points apart. This is not a clear winner —
it is a parallel tracks signal:

1. **Pilot with billing company first.** 60 days. Generates denial rate
   reduction data from real MA + Medicaid billing. Low cost.

2. **Use that evidence to close NikoHealth.** The pilot data makes the
   NikoHealth asset sale conversation 3× easier and likely 2× more valuable
   in price negotiation.

These tracks are mutually reinforcing, not competing.

**VGM is the correct follow-on distribution deal** after the NikoHealth asset
sale or billing company deployment — not the primary target during MA + Medicaid
proof-of-concept.

---

## MA + Medicaid: The Payer Complexity That Changes Everything

```
FFS Medicare: One ruleset per MAC jurisdiction (4 MACs nationally)
Medicare Advantage: 800+ plans, each with plan-specific PA rules
                    layered on top of CMS Part B baseline
Medicaid: 50 state rulesets, often MCO-layered within each state
          TRAP: Several states have moved CGM coverage to pharmacy-only
          (NY May 2024; others following) — verify before piloting
```

**Why this matters for NikoHealth:** Maintaining current MA plan-specific
PA rules across 800+ plans is ongoing data work, not engineering. A supplier
tool that owns this database has a moat that is expensive to replicate.

**Why this matters for billing companies:** They are already maintaining
these rules manually in someone's spreadsheet. The pain is personal and daily.

---

## Key Denial Quantification

| Metric | Value | Source |
|--------|-------|--------|
| CGM improper payment rate | 25.2% | CMS 2024 |
| Projected annual improper payments | $278.5M | CMS 2024 |
| Documentation failure share | 94.2% | CMS 2024 |
| No documentation at all | 67.6% | CMS 2024 |
| Insufficient documentation | 26.6% | CMS 2024 |
| MA DMEPOS appeal success (L2) | 63.9% | KFF 2024 |
| MA PA denial rate | 7.7% | KFF 2024 |
| Write-off rate on denied claims | ~63% | Derived from appeal ladder economics |
| Net revenue loss after appeals | ~20% of gross CGM billing | Derived |

---

## Sources
- [CMS 2024 CGM Improper Payments](https://www.cms.gov/training-education/medicare-learning-networkr-mln/compliance/medicare-provider-compliance-tips/glucose-monitoring-supplies)
- [KFF: MA 53M PA Determinations 2024](https://www.kff.org/medicare/nearly-50-million-prior-authorization-requests-were-sent-to-medicare-advantage-insurers-in-2024/)
- [MA Denial Spike — Medicare Rights Center](https://www.medicarerights.org/medicare-watch/2024/09/26/medicare-advantage-denials-increased-before-the-implementation-of-new-prior-authorization-rules)
- [Unequal DME/Pharmacy Enforcement — CMS Comment](https://downloads.regulations.gov/CMS-2025-0242-0025/attachment_1.pdf)
- [UHC/Synapse Health State Expansion](https://www.uhcprovider.com/en/resource-library/news/2025/synapse-health-manage-dme-orders.html)
- [OIG CGM Payments Exceeded Costs 2025](https://oig.hhs.gov/reports/all/2025/medicare-payments-for-continuous-glucose-monitors-and-supplies-exceeded-supplier-costs-and-retail-market-prices-indicating-medicare-can-save-at-least-tens-of-millions-of-dollars-in-one-year/)
- [NY State Medicaid CGM Billing Update 2024](https://www.emedny.org/ProviderManuals/communications/billing_glucose_monitors_-_5-9-24.pdf)
- [NikoHealth vs Brightree 2026](https://coruzant.com/software/dme-and-hme-software-in-2026/)
ENDDOC
ok "signal-cgm-re-scored-composite-v2.md written"

# =============================================================================
# FILE 63 — signal-cgm-final-ranking-leverage-v3.md
# Source: "Using the latest composite scoring model..." final analysis
# =============================================================================
log "Writing signal-cgm-final-ranking-leverage-v3.md..."
cat > "$ANALYSIS_DIR/signal-cgm-final-ranking-leverage-v3.md" << 'ENDDOC'
# Signal CGM — Final Composite Ranking and Leverage Analysis v3
### Asset Sale Primary · Pilot Secondary · MA + Medicaid Payer Scope
### STTIL Solutions LLC | April 2026

---

## Final Composite Ranking — Four Segments, MA + Medicaid Scope

| Rank | Segment | Asset (×0.50) | Pilot (×0.30) | SaaS (×0.20) | Composite |
|------|---------|--------------|--------------|-------------|-----------|
| **#1** | Billing company / DME RCM | 3.60 | 2.46 | 1.55 | **7.61** |
| **#2** | NikoHealth-type platform | 4.28 | 2.13 | 0.93 | **7.34** |
| **#3** | VGM Group / MSO | 3.58 | 1.08 | 1.06 | **5.72** |
| **#4** | Mid-size independent supplier | 1.65 | 2.09 | 1.62 | **5.36** |

**The 0.27-point gap between #1 and #2 is a sequencing signal, not a clear winner.**
Pilot with billing company → generate evidence → close NikoHealth asset sale at higher price.
These tracks are mutually reinforcing.

---

## Denial Cost Absorption vs. Recovery — Full Quantification

### Appeal Ladder Economics (Per ~$250 CGM Claim)

| Level | Timeline | Staff Cost | Overturn Rate | Net Yield |
|-------|----------|-----------|--------------|-----------|
| L1 Redetermination | 60–74 days | $25–$55 | 20–30% | $22.50 net after labor |
| L2 QIC Reconsideration | 60 days | $50–$118 | 50–65% | $58.50 net after labor |
| L3 ALJ Hearing | 6–18 months | $200–$1,500+ | 60–70% | ($687.50) net LOSS per claim |
| L4–5 Council / Court | Years | $5K–$50K+ | Variable | Not viable for CGM refills |

**MA-specific overlay:** MA L2 overturn rate is 63.9% (vs ~57% FFS) — but MA plans
deny more aggressively initially. Net: higher appeal labor per recovered dollar.

### Absorption Model — 500-Patient Supplier, One Month

```
500 active CGM patients | $125,000/month billing | 25.2% improper payment rate

126 denied claims (~$31,500 face value)
  ├── 38 not appealed → written off immediately        $9,500
  ├── 50 appealed at L1 only
  │    ├── 13 overturned (~27%)                        $3,250 recovered
  │    ├── 25 denied → abandoned                       $6,250 written off
  │    └── 12 escalated to L2
  └── 38 at L2 total (12 escalated + 26 direct)
       ├── 22 overturned (~57%)                        $5,500 recovered
       └── 16 denied → written off                     $4,000 written off

OUTCOME LEDGER:
  Permanently written off (all paths):   79 claims   $19,750   63%
  Recovered through appeals:             35 claims    $8,750   28%
  Still in process:                      12 claims    $3,000    9%

  Appeal labor cost (88 filings):                     $5,192/month
  Net recovery after labor:              $8,750 − $5,192 = $3,558
  Net loss including labor:             $19,750 + $5,192 = $24,942

ONE IN FIVE CGM BILLING DOLLARS IS PERMANENTLY LOST.
The appeal process recovers less than 30 cents of every denied dollar
after accounting for the staff labor required to run it.
```

---

## The Six Situations Where Suppliers Ship Despite Documentation Risk

### Situation 1: The PA-Pending Ship
**What:** PA submitted but not yet affirmed. Refill due in 5 days. Supplier
ships assuming approval will come through.
**Frequency:** Very high — 14–21 day MAC adjudication window routinely overlaps
refill schedule for reactive workflows.
**Outcome if PA denied:** Non-covered denial. Zero recovery path.
**Signal CGM intervention:** PA initiated 45 days out. Order blocked until
PA confirmed. Ship date never collides with adjudication window.

### Situation 2: The CMN Gray Zone
**What:** CMN expired 60 days ago. Doctor's office has been faxed three times.
Insulin-dependent patient is out of supplies. Supplier ships anyway.
**Frequency:** High. Physicians have 200 other patients. CMN renewal takes
60–90 days in slow practices.
**Outcome:** Denial. Backdated CMN rarely accepted at redetermination.
**Signal CGM intervention:** CMN expiration flagged 60 days out. Outreach
triggered at 45 days. Hold queue activates at 30 days if CMN not received.

### Situation 3: The New-Code Gap
**What:** CMS adds codes to Required PA list (7 new codes April 13, 2026).
Staff not notified. Existing workflow processes orders normally. No PA obtained.
**Frequency:** Episodic but acute at each list expansion. Hypothesis 2 in
validation-hypotheses.md tests whether this gap is live and unpatched now.
**Outcome:** Non-covered denial. No recovery.
**Signal CGM intervention:** Required PA code list maintained current and
applied automatically to all open refill windows.

### Situation 4: The PECOS Assumption
**What:** Prescriber was enrolled at intake 18 months ago. Practice changed.
PECOS lapsed. Supplier has no system to re-check at refill cycle.
**Frequency:** Moderate but growing. Practice instability post-COVID increasing.
**Outcome:** Hard denial. No appeal path if prescriber genuinely not enrolled.
**Signal CGM intervention:** NPPES checked at intake AND at each refill cycle.
Inactive NPI → order blocked → alert to supplier staff.

### Situation 5: The Synapse Blindside
**What:** Patient on UHC Medicare Advantage. Valid UHC PA in hand. Unknown to
supplier: UHC transitioned patient's state to Synapse Health (April 1, 2026).
Supplier not enrolled in Synapse. PA obtained through UHC portal not valid
in Synapse-managed network. Order ships. Denied: not in authorized network.
**Frequency:** Acute. 20+ states now in Synapse territory as of April 2026.
**Outcome:** Network access denial. Difficult appeal — supplier must prove
non-notification.
**Signal CGM intervention:** Payer-plan tracking layer flags UHC MA patients
in Synapse-covered states. Verifies Synapse enrollment before shipment.

### Situation 6: The Continuity Bridge
**What:** Coverage lapsed. Supplier working to restore. Patient calls: out of
sensors, insulin-dependent. Supplier provides bridge supplies with no active
PA, no valid CMN, no documentation.
**Frequency:** Low per patient; high consequence. The end state of situations
1–4 going unmanaged.
**Outcome:** Full product cost absorbed. No billing path.
**Signal CGM intervention:** Situations 1–4 never reach Situation 6 when
caught 45 days out. This situation is structurally prevented by the other five.

---

## Workflow Leverage Analysis

| Step | Financial Impact/Event | Frequency/Month | Preventability | Leverage Score |
|------|----------------------|----------------|----------------|----------------|
| **Prior Authorization** | $250 hard write-off; zero recovery path | HIGH: 15–25 at-risk refills in 500-pt book | HIGH: PA tracked 45 days out; ship blocked until confirmed | **9.2 / 10** |
| **Refill Tracking (Coverage Clock)** | Enables all other steps | HIGH: every active patient every month | HIGH: core calendar engine | **8.8 / 10** |
| **6-Month Visit Compliance** | $250/claim; 55% recoverable at L2 | MODERATE: 8–15 patients/month in mature book | HIGH: visit window calculated from refill date | **8.1 / 10** |
| **PECOS Validation** | $250 hard write-off; zero recovery | LOW: 2–5 events/month in 500-pt book | HIGH: NPPES checked at intake + each refill | **7.4 / 10** |
| **Intake Validation** | Prevents pipeline contamination | LOW: 5–15 new patients/month in mature book | HIGH: NPI, eligibility, base equipment, duplicates | **6.5 / 10** |
| **Audit Defense Log** | High if audit triggered ($50K–$500K) | LOW: episodic | MODERATE: time-stamped byproduct of all other steps | **5.8 / 10** |

### The Leverage Verdict

**Prior authorization is the single highest-leverage intervention point.** PA
failure = non-recoverable write-off. No appeal path. Zero exceptions. Every
other denial type has some theoretical recovery path. PA does not.

**Refill tracking is the enabling infrastructure, not a standalone lever.**
Without knowing when the next order ships, nothing else is actionable. It is
the architecture, not a feature.

**6-month visit compliance is the highest-frequency daily queue driver.** 8–15
patients per month, every month, in a 500-patient mature book. Prevented before
shipment is worth more than 57% chance of recovery after denial.

**Audit defense is the second story for asset sale conversations.** After daily
denial prevention, before competitive bidding urgency. In the current OIG
enforcement environment, "your customers can prove proactive compliance" is a
meaningful product differentiator.

---

## Asset Sale One-Liner

> "Signal CGM gives the buyer's customers 45 days to fix what would otherwise
> cost them 20% of their CGM revenue — and generates a compliance record that
> defends them against the government enforcement environment that's already active."

---

## Sources
- [CMS 2024 CGM Improper Payments — 25.2% rate](https://www.cms.gov/training-education/medicare-learning-networkr-mln/compliance/medicare-provider-compliance-tips/glucose-monitoring-supplies)
- [VGM DMEPOS Appeals Guide](https://www.vgm.com/communities/navigating-dmepos-appeals-processes-a-strategic-guide-for-providers/)
- [Denial Management Metrics](https://www.panahealthcaresolutions.com/blogs/denial-management-metrics-for-faster-reimbursement/)
- [Experian State of Claims 2025](https://www.experian.com/blogs/healthcare/state-of-claims-2025/)
- [KFF: MA Prior Authorization 2024](https://www.kff.org/medicare/nearly-50-million-prior-authorization-requests-were-sent-to-medicare-advantage-insurers-in-2024/)
- [OIG CGM Enforcement Priority 2025](https://oig.hhs.gov/reports/all/2025/medicare-payments-for-continuous-glucose-monitors-and-supplies-exceeded-supplier-costs-and-retail-market-prices-indicating-medicare-can-save-at-least-tens-of-millions-of-dollars-in-one-year/)
- [UHC/Synapse Health Expansion — UHCprovider.com](https://www.uhcprovider.com/en/resource-library/news/2025/synapse-health-manage-dme-orders.html)
- [CGM Fraud Schemes — Qlarant](https://www.qlarant.com/knowledge/blog/continuous-glucose-monitor-fraud-schemes/)
- [Ossur Medicare Appeals Timeline Guide](https://www.ossur.com/en-us/professionals/ossur-rr/appealing-denied-medicare-claims-redetermination-reconsideration-and-alj)
ENDDOC
ok "signal-cgm-final-ranking-leverage-v3.md written"

# =============================================================================
# PITCH v1 — signal-cgm-pitch-v1-plain.md (Plain Language / Social Media)
# =============================================================================
log "Writing signal-cgm-pitch-v1-plain.md..."
cat > "$ASSETS_DIR/signal-cgm-pitch-v1-plain.md" << 'ENDDOC'
# Signal CGM — Asset Sale Overview
## Plain Language Edition
### For Social Sharing and Non-Technical Decision Makers

---

> SIGNAL CGM
> Stop Losing 20% of Your Clients' CGM Revenue.
> A ready-to-deploy tool for DME billing companies.
> STTIL Solutions LLC | kisasttil@gmail.com

---

### The Problem. In Plain English.

Right now, your clients are losing money they don't have to lose.

For every $100 they bill on CGM (continuous glucose monitors), about **$20
disappears.** It doesn't go to fraud. It doesn't go to bad patients. It goes
to paperwork that wasn't ready on time.

A claim gets denied. The product was already shipped. The patient has it. Your
client can't get it back. They try to appeal. Most of the time, they lose more
money fighting it than they get back.

Here's the part that stings: **94% of those denied claims could have been
prevented.** The information existed. Someone just didn't have it in hand
before the order shipped.

That's what Signal CGM fixes.

---

### What Goes Wrong — and When We Catch It

There are six places in the workflow where a claim goes from "fine" to "denied."
We catch all six **before the product ships**, not after.

**1. Prior Authorization — the biggest one.**
No PA before shipment = no money, period. No appeals, no second chances. We
make sure PA is confirmed 45 days before the order goes out the door.

**2. The Coverage Clock.**
Every CGM patient has a refill schedule. We track it for every patient, every
month. This is the engine everything else runs on.

**3. The 6-Month Doctor Visit.**
Medicare requires a check-in with the prescribing doctor every 6 months for
CGM patients to keep getting supplies. When that visit doesn't happen, the
claim gets denied. We flag it a month before it becomes a problem.

**4. Prescriber Enrollment Check.**
If the doctor who ordered the CGM isn't currently enrolled in Medicare, the
claim gets denied — even if everything else is perfect. We re-check this every
time an order is about to ship, not just when the patient first signs up.

**5. New Patient Setup.**
Before anyone's first order ships, we check eligibility, duplicate claims, and
whether the right supplier is on file with CMS. Bad setups become expensive
surprises later.

**6. Audit Defense — the safety net.**
Every check we run gets logged with a time stamp. If CMS ever audits one of
your clients, that log shows exactly what was verified and when. It's proof
they were doing things right.

---

### Try It First. On Your Own Client Data.

We're offering a **60-day pilot** at no cost.

Pick two to three of your suppliers. We run Signal CGM on their live CGM
patient data. We track what would have been denied. We show you what changed.

At the end of 60 days, you'll see the before-and-after in your clients' actual
numbers — not in a demo, not in a made-up scenario.

If it doesn't show a clear improvement in first-pass CGM claims, there's no deal.

---

### What We're Asking

Signal CGM is available two ways:

**Option 1 — You own it.**
One-time purchase: **$45,000–$65,000.**
You get the full code, all the research, 30 days of live handoff sessions, and
the AI development context so your team can keep building. You white-label it.
You charge your clients. The revenue is yours.

**Option 2 — Per-client licensing.**
**$75 per supplier client per month.**
If you have 20 CGM-active clients, that's $1,500/month. You pass the cost
through at whatever margin makes sense for your business.

---

### Three Questions You're Probably Already Thinking

**"Is this worth the cost?"**
Your clients are losing $20 of every $100 they bill on CGM. If Signal CGM
moves that to $14 lost instead of $20 — a modest improvement — a single
500-patient supplier recovers $9,000+ per month. At $75/month, that's a
120-to-1 return. The cost question answers itself.

**"Doesn't this make my billing services less necessary?"**
No. Your clients still need you to run their billing, handle denials, and
manage payer relationships. Signal CGM handles the pre-shipment window — the
45 days before a claim exists. That's not your current job. It becomes a new
service you offer, not a replacement for what you already do.

**"What happens to patient data?"**
Signal CGM never stores patient names, Social Security numbers, dates of birth,
or contact information. The only identifier the system uses is the supplier's
internal patient ID number. All audit logs hash even that. Data stays on your
infrastructure, not ours. A Business Associate Agreement is part of every
deployment.

---

**Ready to run the pilot?**
Contact: kisasttil@gmail.com
STTIL Solutions LLC | Signal CGM

---

*Key data: CMS 2024 CGM improper payment rate 25.2% / $278.5M projected annual.*
*Source: [CMS Glucose Monitoring Compliance](https://www.cms.gov/training-education/medicare-learning-networkr-mln/compliance/medicare-provider-compliance-tips/glucose-monitoring-supplies)*
ENDDOC
ok "signal-cgm-pitch-v1-plain.md written"

# =============================================================================
# PITCH v2 — signal-cgm-pitch-v2-professional.md (Professional)
# =============================================================================
log "Writing signal-cgm-pitch-v2-professional.md..."
cat > "$ASSETS_DIR/signal-cgm-pitch-v2-professional.md" << 'ENDDOC'
# Signal CGM — Asset Sale Overview
## Professional Edition
### Strategic Acquisition for DME-Focused Revenue Cycle Management Organizations

---

> SIGNAL CGM
> A CGM-Specific Denial Prevention and Compliance Intelligence Platform
> for DME Revenue Cycle Management Organizations
>
> Asset Acquisition and Licensing Opportunity | April 2026
> STTIL Solutions LLC | kisasttil@gmail.com

---

### The Problem Your Clients Cannot Solve Alone

CMS's 2024 Medicare Fee-for-Service data puts the CGM improper payment rate at
**25.2% — $278.5 million in projected annual overpayments** on glucose monitor
billing. Of that figure, 94.2% traces directly to documentation deficiencies:
missing or expired Certificates of Medical Necessity, prior authorizations not
obtained before shipment, undocumented 6-month physician visits, and prescriber
PECOS enrollment failures. Less than 6% reflects medical necessity disputes or
coverage policy conflicts.

The financial consequence for a supplier billing 500 active CGM patients monthly
runs deeper than the denial rate suggests. After accounting for appeal labor,
write-off timing, and the hard limits of the five-level Medicare appeals process,
**approximately 63% of denied CGM claim value is permanently absorbed** — not
recovered through redetermination or QIC reconsideration. The net effect on a
$125,000/month CGM billing book: a structural revenue loss of roughly $25,000
per month, or **20% of gross CGM billing**, that does not appear as a line item
in any supplier's P&L but is reflected across AR aging, appeal staffing overhead,
and uncompensated product cost.

This is the problem Signal CGM was built to eliminate — not by improving the
appeals process, but by making the appeals process unnecessary.

---

### Six Workflow Leverage Points, in Priority Order

**1. Prior Authorization Tracking (Highest Leverage — No Recovery Path if Missed)**
PA not obtained before CGM shipment is the only denial type with zero appeal
recovery. Signal CGM initiates PA tracking 45 days before the projected refill
date — sufficient runway for the MAC's 14–21 day adjudication window plus a
resubmission buffer — and blocks order release until PA is confirmed in the
system. The April 13, 2026 expansion of the Required Prior Authorization list
and ongoing MA plan-specific requirements (UHC non-T1D since September 2024)
are automatically reflected in the current HCPCS code tracking layer.

**2. Refill Tracking / Coverage Clock (Enabling Architecture)**
A predictive refill calendar keyed to each patient's last dispense date, device
type, and payer-specific wear-day rules generates the operational foundation for
every other intervention point. Without forward visibility into the refill
schedule, PA initiation, CMN flagging, and visit compliance checks are reactive
lookups rather than automated workflow triggers.

**3. 6-Month Physician Visit Compliance (Highest Frequency Preventable Denial)**
Medicare's continued CGM coverage requirement mandates a documented in-person
or telehealth visit with the treating practitioner every six months. In a mature
500-patient book, 8–15 patients per month are approaching or past this window.
Unlike PA failures, missed-visit denials carry a 50–65% QIC overturn rate when
documentation can be obtained retroactively — but preventing the denial is worth
more than recovering half of it post-filing. Signal CGM surfaces each patient's
visit window as a prioritized outreach task 30+ days before the refill date.

**4. Prescriber PECOS Validation at Each Refill Cycle (Hard Write-Off Prevention)**
Medicare requires that the ordering physician maintain active enrollment at the
time each order is placed — not merely at the time of initial patient intake.
No incumbent DME billing platform currently re-validates prescriber enrollment
status at the refill cycle level. Signal CGM queries the NPPES registry against
the ordering provider's NPI at each scheduled refill and routes any inactive or
mismatched NPI to a supplier alert queue before the order can release.

**5. Intake Validation (Pipeline Defense)**
At new patient intake, Signal CGM performs eligibility verification, duplicate
claim history check, base equipment record validation (M124), and initial
prescriber PECOS status check before any first order is authorized.

**6. Audit Defense Log (Compliance Record as System Byproduct)**
Every pre-submission check generates a time-stamped audit log entry documenting
what was verified and when. In the current enforcement environment — with $1.9B
in DMEPOS improper payments under active OIG scrutiny, CGM explicitly identified
as a 2026 nationwide enforcement priority, and $1.8B in payments suspended by
CMS's Fraud Defense Operations Center in 2025 — a defensible compliance record
distinguishes a legitimate supplier from an audit target when the MAC issues an ADR.

---

### The Pilot Offer: 60-Day Proof of Concept on Live Client Data

We are offering a structured 60-day pilot at no cost or obligation.

Select two to three CGM-active clients from your book. Signal CGM runs against
their live billing data — tracking open refills, flagging PA status gaps, CMN
expirations, and visit compliance windows across their active CGM patient roster.
At day 30 and day 60, we deliver a denial risk exposure report: claims that would
have shipped without documentation in hand, segmented by denial type and estimated
dollar exposure.

The pilot does not require replacing or integrating with existing billing software.
Signal CGM operates as a parallel layer over whatever clearinghouse or platform
the client currently uses. PHI handling during the pilot is governed by a Business
Associate Agreement executed before data access begins.

At the conclusion of 60 days, you hold a documented, client-specific before-and-after
comparison. That evidence either supports a deployment decision or it does not.
There is no obligation if the results do not meet your threshold.

---

### The Ask

**Option 1 — Full Asset Acquisition**
One-time acquisition: **$45,000–$65,000**

Includes: full Python/FastAPI source code, coverage calculator, audit logger,
PostgreSQL data models, payer rules configuration, complete research library
(market research, compliance roadmap, BAA framework), 30-day live knowledge
transfer with the STTIL Solutions founder, and the CLAUDE.md AI development
context enabling immediate continuation of development with Claude Code at
zero ramp-up cost. No licensing fees, no royalties, no ongoing STTIL involvement
unless contracted separately.

Buyer white-labels the product, deploys across their client base, and captures
the full downstream revenue.

**Option 2 — Per-Seat Licensing**
**$75 per supplier client per month** (volume negotiable above 25 seats)

Includes ongoing payer rule updates, Required PA code list maintenance, and
access to product improvements. A 20-client deployment at $75/month represents
$1,500/month in licensing cost against a conservative $9,000–$18,000/month in
recovered denial revenue across those clients — before accounting for staff
labor savings on appeal management.

---

### Three Objections Addressed Directly

**"The cost isn't justified at our current scale."**
The financial threshold for ROI is low. A single client with 300 active CGM
patients billing $75,000/month, running at the documented 25% improper payment
rate, absorbs approximately $11,200/month in net CGM write-offs after exhausting
the appeal process. If Signal CGM prevents 40% of those — the conservative end
of what pre-submission PA and CMN tracking demonstrably delivers — that client
recovers $4,500/month. Against a $75/month per-seat cost, that is a 60-to-1
return on a single deployment. The 60-day pilot makes this calculation concrete
on your clients' actual numbers.

**"This tool reduces the complexity that makes our billing services valuable."**
Signal CGM operates exclusively in the pre-submission window — the 45 days before
a claim exists. It does not touch denial management, appeals coordination, payer
negotiation, remittance reconciliation, or any other function that defines your
current service relationship. What it does is give your clients fewer denials to
manage, which reduces the reactive workload on your team without displacing any
service functions you bill for. The more accurate framing: Signal CGM converts
reactive denial management — which your staff absorbs at $50–$118 per appeal
cycle — into a proactive workflow your clients pay you a premium to maintain.
It is a service tier expansion, not a service substitution.

**"We can't expose client PHI to a third-party system."**
Signal CGM was designed from the architecture level with this constraint as
non-negotiable. The system ingests five fields only: patient ID (the supplier's
internal account number, not a Medicare beneficiary identifier), device type,
shipment date, quantity, and payer code. No patient names, dates of birth, Social
Security numbers, diagnoses, or contact information enter the system at any point.
All audit logs hash even the patient ID before storage. The system is self-hosted
— it runs on your infrastructure or your client's infrastructure, not on STTIL's
servers. Data never transits a third-party network. The Business Associate
Agreement and full compliance documentation package are included in both
acquisition and licensing structures.

---

**Next Step: Schedule the pilot conversation.**
kisasttil@gmail.com | STTIL Solutions LLC | Signal CGM

*This document is a confidential business communication intended for the named
recipient only.*

---

### Sources
- [CMS 2024 CGM Improper Payment Rate — 25.2% / $278.5M](https://www.cms.gov/training-education/medicare-learning-networkr-mln/compliance/medicare-provider-compliance-tips/glucose-monitoring-supplies)
- [OIG 2025: CGM Payments Exceeded Supplier Costs](https://oig.hhs.gov/reports/all/2025/medicare-payments-for-continuous-glucose-monitors-and-supplies-exceeded-supplier-costs-and-retail-market-prices-indicating-medicare-can-save-at-least-tens-of-millions-of-dollars-in-one-year/)
- [Federal Authorities Targeting CGM Claims — National Law Review](https://natlawreview.com/article/federal-authorities-are-targeting-continuous-glucose-monitoring-cgm-device)
- [MA Prior Authorization Denial Rates — KFF 2024](https://www.kff.org/medicare/nearly-50-million-prior-authorization-requests-were-sent-to-medicare-advantage-insurers-in-2024/)
- [UHC/Synapse Health DME Orders — UHCprovider.com](https://www.uhcprovider.com/en/resource-library/news/2025/synapse-health-manage-dme-orders.html)
- [CMS Prior Authorization Expansion April 13, 2026](https://www.hcintellect.com/post/medicare-dmepos-prior-authorization-expansion-effective-april-13-2026)
- [Denial Management Metrics](https://www.panahealthcaresolutions.com/blogs/denial-management-metrics-for-faster-reimbursement/)
ENDDOC
ok "signal-cgm-pitch-v2-professional.md written"

# =============================================================================
# MASTER SUMMARY
# =============================================================================
log "Writing master-summary.md..."
cat > "$OUTPUT_DIR/master-summary.md" << 'ENDDOC'
# Signal CGM — Master Summary
### Strategic Analysis Package | STTIL Solutions LLC | April 2026

---

## What This Package Contains

| File | Description |
|------|-------------|
| Analysis/signal-cgm-segment-scoring-v1.md | Three-model scoring: SaaS ICP, Asset Sale, Pilot Validation across 7 segments |
| Analysis/signal-cgm-re-scored-composite-v2.md | Composite re-score (50/30/20 weights) with MA + Medicaid payer scope applied |
| Analysis/signal-cgm-final-ranking-leverage-v3.md | Final four-segment ranking, denial quantification, six leverage points |
| Assets/signal-cgm-pitch-v1-plain.md | Leave-behind: plain language / social media (6th grade reading level) |
| Assets/signal-cgm-pitch-v2-professional.md | Leave-behind: professional (mid-size and larger billing organizations) |

---

## Final Composite Rankings (MA + Medicaid Scope · Asset Sale Primary)

| Rank | Segment | Composite Score | Primary Use |
|------|---------|----------------|-------------|
| #1 | Billing company / DME RCM outsourcer | **7.61** | Pilot partner + asset buyer |
| #2 | NikoHealth-type platform vendor | **7.34** | Primary asset sale target |
| #3 | VGM Group / MSO | **5.72** | Follow-on distribution post-sale |
| #4 | Mid-size independent supplier | **5.36** | SaaS ICP if pivot away from asset sale |

**Weights:** Asset sale 50% · Pilot 30% · SaaS 20%

---

## Key Metrics (All Sources: CMS 2024 / OIG 2025 / KFF 2024)

| Metric | Value |
|--------|-------|
| CGM improper payment rate (Medicare) | 25.2% |
| Projected annual CGM improper payments | $278.5M |
| Share from documentation failures | 94.2% |
| Net revenue loss after appeals (per supplier) | ~20% of gross CGM billing |
| Permanently written off (not recovered) | ~63% of denied claim value |
| Recovered through L1+L2 appeals | ~28% of denied claim value |
| MA DMEPOS appeal success rate (L2) | 63.9% |
| MA prior auth denial rate | 7.7% of PA requests |
| Traditional HME supplier locations (2024) | ~8,005 (−38% from 2013) |

---

## Leverage Priority Stack (Workflow Intervention Points)

1. **Prior Authorization** (9.2/10) — Only denial type with zero recovery path
2. **Refill Tracking / Coverage Clock** (8.8/10) — Enabling infrastructure for everything
3. **6-Month Visit Compliance** (8.1/10) — Highest-frequency daily queue driver
4. **PECOS Validation at Each Refill** (7.4/10) — Hard write-off prevention
5. **Intake Validation** (6.5/10) — Front-door pipeline defense
6. **Audit Defense Log** (5.8/10) — Compliance record as system byproduct

---

## Sequencing Recommendation

```
Week 1–2:   Identify 2–3 billing company targets with CGM-active client books
Week 2–4:   Approach with Version 1 or 2 leave-behind; propose 60-day pilot
Week 4–8:   Execute pilot on live MA + Medicaid CGM data under BAA
Week 8:     Deliver denial risk exposure report (before/after)
Week 9–12:  Use pilot evidence to open NikoHealth asset sale conversation
            at higher valuation than current $45K–$65K ask
Week 12+:   Parallel VGM vendor partner program conversation for distribution
```

---

## Validated Hypotheses Status (as of April 2026)

| Hypothesis | Status | Method |
|-----------|--------|--------|
| H1: Denials are documentation-fixable | **Confirmed by CMS data** — 94.2% doc failures | Desk research (OIG + CMS 2024) |
| H2: April 13 PA expansion is live and unpatched | **Likely confirmed** — no incumbent updated | Requires 2–3 discovery calls to verify |
| H3: CB 2028 deadline drives active buying urgency | **Untested** | Requires 5 discovery calls with owner-operators |

---

## Next-Steps Prompt for Next Claude Code Session

> Continue Signal CGM go-to-market execution. All strategic analysis is
> in CGM-Denial-Prevention/01-Claude-Outputs/. The asset sale target ranking
> is: #1 Billing company (pilot first), #2 NikoHealth (asset sale after pilot
> evidence). Payer scope: Medicare Advantage and Medicaid as first
> proof-of-concept. Pilot offer: 60 days, no cost, on live client data under
> BAA. Asset ask: $45K–$65K acquisition or $75/client/month licensing.
> Next priority: identify 3 billing company targets (suggest starting with
> Florida-based DME RCM firms given existing FAHCS research) and prepare
> outreach sequence using Assets/signal-cgm-pitch-v2-professional.md.
> Hypothesis H2 (April 13 PA gap) and H3 (CB urgency) still require
> discovery call validation per validation-hypotheses.md in Obsidian vault.

---

*Generated: April 2026 | STTIL Solutions LLC | Signal CGM*
*All figures from CMS, OIG, KFF primary sources — see individual files for citations*
ENDDOC
ok "master-summary.md written"

# =============================================================================
# Step 6: Git Operations
# =============================================================================
echo ""
echo "  ── Git Operations ──────────────────────────────────────────────"
cd "$REPO_ROOT"

# Verify git repo
git rev-parse --git-dir > /dev/null 2>&1 || fail "Not a git repository: $REPO_ROOT"

# Stage the new directory
log "Staging CGM-Denial-Prevention/..."
git add CGM-Denial-Prevention/

# Check if there is anything to commit
if git diff --cached --quiet; then
  ok "Nothing new to commit — files already tracked and unchanged"
else
  log "Committing..."
  git commit -m "$(cat <<'MSG'
feat: Signal CGM strategic analysis + asset sale package

Adds complete go-to-market analysis for Signal CGM asset sale:

Analysis/
  - signal-cgm-segment-scoring-v1.md   (3-model scoring across 7 segments)
  - signal-cgm-re-scored-composite-v2.md (50/30/20 composite, MA+Medicaid scope)
  - signal-cgm-final-ranking-leverage-v3.md (final 4-segment rank + leverage map)

Assets/
  - signal-cgm-pitch-v1-plain.md       (plain language leave-behind)
  - signal-cgm-pitch-v2-professional.md (professional leave-behind)
  - master-summary.md                   (rankings, metrics, next-steps prompt)

Key findings: 25.2% CGM improper payment rate; 20% net revenue loss;
63% of denied claims permanently written off; billing company #1 target
for pilot; NikoHealth #1 for asset sale.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
MSG
)"
  ok "Committed"

  # Push
  log "Pushing to origin..."
  if git push 2>&1; then
    ok "Pushed to origin"
  else
    echo "  ⚠ Push failed — changes are committed locally. Run 'git push' manually."
    echo "    (This is non-fatal; all files are written and committed.)"
  fi
fi

# =============================================================================
# Step 7: Obsidian Sync
# =============================================================================
echo ""
echo "  ── Obsidian Sync ───────────────────────────────────────────────"

if [ -d "$(dirname "$OBSIDIAN_DIR")" ]; then
  log "Creating Obsidian target directory..."
  mkdir -p "$OBSIDIAN_DIR"

  log "Syncing Analysis files..."
  cp "$ANALYSIS_DIR/"*.md "$OBSIDIAN_DIR/"

  log "Syncing Assets files..."
  cp "$ASSETS_DIR/"*.md "$OBSIDIAN_DIR/"

  log "Syncing master-summary.md..."
  cp "$OUTPUT_DIR/master-summary.md" "$OBSIDIAN_DIR/"

  ok "Obsidian vault synced → $OBSIDIAN_DIR"
else
  echo "  ⚠ Obsidian vault path not found: $(dirname "$OBSIDIAN_DIR")"
  echo "    Skipping Obsidian sync — all files are in $OUTPUT_DIR"
fi

# =============================================================================
# Done
# =============================================================================
echo ""
echo "━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━"
echo "✅ All assets captured"
echo ""
echo "  Output directory : $OUTPUT_DIR"
echo "  Obsidian sync    : $OBSIDIAN_DIR"
echo ""
echo "  Files written:"
echo "    Analysis/signal-cgm-segment-scoring-v1.md"
echo "    Analysis/signal-cgm-re-scored-composite-v2.md"
echo "    Analysis/signal-cgm-final-ranking-leverage-v3.md"
echo "    Assets/signal-cgm-pitch-v1-plain.md"
echo "    Assets/signal-cgm-pitch-v2-professional.md"
echo "    master-summary.md"
echo "━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━"