# Asset Nova — Strategy Decision & Long-Term Roadmap

Status: working draft for case study. Sources: `context-library/research/Orbitron/cpm-agentic-ai-product-discovery.md` + `outputs/analyses/orbitron-context-synthesis.md`. Date: 2026-08-26.

## Strategic thesis (agreed)

- Every AI competitor in capital projects is construction-phase, single-silo, contractor-side. The open seat: **owner-side, cross-system, full-lifecycle intelligence** for process-industry megaprojects. Middle East = worst pain globally (89% over cost, 87% late) = Orbitron's home turf.
- The real product is the **substrate**: a cross-system, document-grounded project record (contracts, correspondence, schedules, costs, registers). Agents are composable skills on top — pure NovaOS architecture, "skills compound, dashboards decay."
- Asset Nova composes existing Orbitron agents: ComplyNova (evidence packs, case routing, audit) and FlowNova (procurement, expediting, contracting, control tower).

## Wedge decision (Hormozi value equation)

**First product: Shield** ("claims defense agent") — reads contracts + live project record continuously, detects incoming contractor claims and variation requests, finds weaknesses in the contractor's position (missed notice windows, weak entitlement basis, unsupported quantum), and helps the owner defend their profit. Human approves defense strategy and response.

Why it wins: dollar-denominated dream outcome (33% of capex disputed, $60M avg dispute) × provable likelihood (document evidence chains exposing contractor position gaps) ÷ days-to-value (runs on historical files) × minimal effort (read-only document ingestion, no ERP integration, lands within one project team's buying authority).

Deprioritized as wedge (kept in roadmap): budget estimation (needs outcome database that doesn't exist yet) and compliance DD (ComplyNova already covers the GRC flavor; CPM-specific compliance folds into Shield + later phases).

## Roadmap — three horizons

### Horizon 1: Land (months 0–6) — "Defend the owner's profit"
- **Shield** (owner-defense: detects incoming contractor claims, identifies weaknesses in contractor positions, drafts defense responses)
- **Schedule Interrogator** — forensic XER diffing + cross-examination against procurement/correspondence; "10 questions to ask the contractor this month"
- Deployment: in-region VPC/sovereign, document-based, paid pilot on 1–2 live projects, success metric = $ claims deflected or reduced
- Substrate v1 built as a byproduct: contracts + correspondence + schedules indexed and linked

### Horizon 2: Expand (months 6–18) — "Run the monthly grind"
- **Design Gate** — AI validation of FEED revisions against building codes, cost impacts, and buildability; catches design drift before it becomes cost drift. Daily-retention workhorse; integrations deepen the moat
- **Deliverable Chaser** — ranks overdue documents by critical-path impact, pre-reviews revisions, drafts chase correspondence
- Substrate v2: cost + document register joined to the record; agent skills registered/versioned in NovaStudio

### Horizon 3: Differentiate (months 18+) — "Gatekeeper of capital"
- **Budget Red Team** — audits front-end definition packages, reference-class forecasting at FID against the owner's own portfolio history ("your productivity assumption exceeds the P90 of comparable Gulf projects"). No competitor touches FEL. Speaks to boards. Customer's own history = compounding data moat ("keep your alpha")
- **Handover Assembler** — tag-level completeness auditing, vendor dossier classification; bridge into O&M/digital twin market and future brownfield projects
- Asset Nova becomes the capital-allocation gatekeeper, closing the full lifecycle loop (each finished project trains the next Budget Red Team review)

## Commercial model
- Outcome-anchored pricing (industry anchor: 0.1–0.3% of project value → 8-figure potential on $10B programs), consistent with Orbitron's "ROI guaranteed" motion
- MENA-first: IKTVA localization, SACS-002 cybersecurity cert, in-Kingdom data residency, pilot below mega-procurement thresholds

## Case study demo plan (36h)
1. Live prototype: Shield detects a $40M incoming contractor claim, identifies 3 weaknesses in the contractor's position (missed notice window, unsupported quantum, contradicted by correspondence), drafts defense response with evidence chain, human approves
2. One orchestration scene: Shield's defense analysis triggers ComplyNova (evidence pack) + FlowNova (downstream impact) in a NovaStudio-style canvas
3. Deck narrative: industry pain → structural insight (why software never fixed it) → substrate thesis → wedge → 3-horizon roadmap → GTM → governance answers (agent owner, scope, escalation, value metric) → 90-day PM plan