Business Case: Legionis

Document ID: DOC-2026-020 Owner: BizOps Date: 2026-02-14 Status: Draft for Review Product: Legionis (legionis.ai) V2V Phase: Phase 2 - Strategic Decisions Related: DR-2026-001 (Pricing Model), DR-2026-003 (Legionis Naming/Repositioning)

Executive Summary

Legionis is an AI workforce platform that deploys teams of specialized, autonomous AI agents to augment human work across any professional domain. Built on the proven Product Org OS foundation (39 agents, 61 skills, 8 gateways, V2V framework), Legionis expands from product management into a general-purpose platform where each professional function becomes a deployable "domain pack."

The Opportunity: Knowledge workers spend the majority of their time on structured professional work that benefits from specialist perspectives, organizational memory, and cross-functional collaboration. Legionis provides pre-built expert teams that deliver these capabilities on demand, at a fraction of the cost and time of assembling human teams or building custom agent frameworks.

The Model: Usage-based pricing with domain pack expansion. Users pay per agent interaction, not per feature gate. Each new domain pack (Product, Engineering, Design, Marketing, Sales, Finance, Legal) opens a new revenue stream on shared infrastructure. Infrastructure costs are fixed at approximately $162/month (validated architecture), with near-zero marginal cost per user thanks to BYOT (Bring Your Own Token) and Google Drive storage.

Why Now:

• The AI agent market is rapidly maturing, with enterprises actively seeking agent-based solutions for professional work
• Product Org OS v3.0 has demonstrated the viability of multi-agent teams (39 agents operating cohesively)
• BYOT architecture eliminates the API cost variability that makes most AI SaaS businesses unprofitable
• The "domain pack" model creates a scalable expansion path that pure-play AI tools cannot replicate
• No competitor offers pre-built, expert-level agent teams across multiple professional domains

1. Problem Statement / Opportunity

The Problem

Knowledge workers face a force multiplication gap:

The gap: There is no platform that provides pre-built teams of domain-expert AI agents that:

• Understand specific professional workflows deeply (not generic prompts)
• Maintain organizational memory that compounds across interactions
• Offer multi-perspective analysis (agents debate and synthesize, not echo)
• Scale across professional domains without starting from scratch
• Respect user data sovereignty and cost transparency

The Opportunity

Product Org OS has proven the model works. 39 specialized agents collaborating through 8 gateways, 61 skills across 6 V2V phases, organizational memory that compounds. The architecture is general-purpose. The Product Org was the first domain pack. The platform can host many more.

What Legionis unlocks:

• For individual professionals: Deploy a full expert team for $[TBD]/month instead of hiring consultants or building custom tools
• For teams: Augment human expertise with AI specialists, maintaining shared organizational memory
• For organizations: Standardize professional workflows across functions with consistent methodology and institutional knowledge

Key insight: Each new domain pack expands SAM without proportional cost increase. The platform infrastructure is shared; only agent definitions, skills, and knowledge packs are domain-specific.

Comparison to prior market sizing: The original business case scoped TAM at ~$284M (300K PMs x $79/mo). By repositioning as a general AI workforce platform, the addressable market expands by an order of magnitude. Product management becomes one of 7+ addressable domains, each with its own professional population measured in hundreds of thousands to millions.

2. Proposed Solution

Product Vision

Legionis: Your AI Workforce. Teams of specialized agents, deployed on demand, for any professional domain.

Latin: "of the legion" - many agents acting as one.

Core Architecture

+------------------------------------------------------------------+

LEGIONIS PLATFORM


Web Interface
Domain Pack: Product Org

* Agent roster
Domain Pack: [Engineering]

* Workspace
[TBD agents, skills, gates]

[Scaffolded: 6 agents]

Domain Pack: [Marketing]

[Scaffolded: 14 agents]


SHARED PLATFORM LAYER

 Agent orchestration     Context layer (org memory)

 Prompt compilation      Cross-team routing

 File operations         Usage tracking & billing

 Gateway orchestration   Multi-tenant isolation (RLS)


USER'S OWN INFRASTRUCTURE

* LLM API key (BYOT - Anthropic, OpenAI, etc.)

* Google Drive storage (documents, context)


+------------------------------------------------------------------+

What Users Get

What Makes This Different

3. Business Model

Revenue Model: Usage-Based with Domain Pack Expansion

Per DR-2026-001: Usage-based pricing. Charge per agent interaction (operation), not per feature gate.

Revenue expansion levers:

Domain Pack Economics

The key insight: domain packs are high-margin, low-cost revenue expansion.

Result: Each new domain pack is essentially pure content investment with near-zero marginal infrastructure cost. Once created, every user who activates it generates revenue at platform-level margins.

Why Usage-Based Changes Everything

Traditional vertical SaaS:

Revenue per user:     $79-149/month (fixed)
COGS per user:        $15-50/month (variable, unpredictable)
Expansion revenue:    Upsell to higher tier (limited)
New market entry:     Build a new product

Legionis (Usage-Based + Domain Packs):

Revenue per user:     Scales with usage (uncapped upside)
COGS per user:        Near-zero (BYOT + Google Drive)
Expansion revenue:    Add domain packs (same user, more revenue)
New market entry:     Add a domain pack (shared infrastructure)

The compounding effect: More domain packs attract more users. More users generate more operations. More operations fund more domain pack development. Each cycle strengthens the platform.

4. Cost Structure

Infrastructure Costs (Validated)

Total: ~$162/month (from Architecture Plan V3 cost analysis, validated Feb 13, 2026)

Lean MVP option: ~$64/month (defer Typesense, Sentry paid, Better Stack)

What We Do NOT Pay For

Infrastructure Scaling Path

Key structural advantage: With BYOT and Google Drive, our costs scale with INFRASTRUCTURE usage (database queries, search indexes), not with USER ACTIVITY (LLM calls, file storage). This decouples cost growth from usage growth.

Cost Per Domain Pack (Incremental)

Observation: Design (6 agents), Architecture (6 agents), and Marketing (14 agents) Extension Teams are already scaffolded in the Product Org OS v3.0. Converting these to Legionis domain packs requires adaptation, not creation from scratch.

5. Unit Economics Framework

Revenue Per User (Framework)

Cost Per User

Gross Margin Analysis

The BYOT advantage is structural: By having users bring their own LLM API keys, we eliminate the single largest and most variable cost in AI SaaS. Every user is profitable from interaction #1.

Lifetime Value Framework

LTV = (Monthly ARPU x Gross Margin) / Monthly Churn Rate
Variables to measure post-launch:
Monthly ARPU: [TBD - usage-based, will vary by user segment]
Gross Margin: ~90-95% (BYOT model)
Monthly Churn: [TBD - target 3-5%]

Expansion LTV multiplier:
Users who add domain packs increase ARPU without increasing acquisition cost
Net Revenue Retention target: >110% (expansion exceeds churn)

Customer Acquisition Cost Framework

CAC:LTV Ratio

Why we expect favorable CAC:LTV:

6. Competitive Advantage

The Moat: Pre-Built Expert Teams + Compounding Organizational Memory

Competitive Positioning

Growth Strategy: Land and Expand

LAND: Product Org pack

  +-- User deploys Product Org agents
  |   (PM, VP Product, BizOps, CI, PMM...)

  +-- Organizational memory accumulates
  |   (Decisions, bets, learnings compound)

  +-- Natural cross-domain needs emerge
  |   ("I need @ui-designer for this PRD")
  |   ("Marketing needs to see this positioning")

EXPAND: Add domain packs

  +-- Design Studio pack
  +-- Marketing Brigade pack
  +-- Engineering Legion pack
  +-- Sales pack
  +-- Finance pack
  +-- Legal pack

  +-- Each pack: new users from that domain
  +-- Each pack: expansion revenue from existing users
  +-- Each pack: cross-domain value increases

COMPOUND: Platform network effects

  +-- More domains = more cross-domain value
  +-- More users = better agent training data (anonymized)
  +-- More context = deeper organizational memory
  +-- More packs = higher switching costs

7. Revenue Model Structure

MRR Growth Framework

Revenue = (Active Users x Avg Operations x Price per Operation) + (Domain Pack Fees)

Growth levers:

ARR Milestone Framework

Note: Subscriber counts depend on final pricing structure (usage-based). The above milestones are revenue-based rather than subscriber-based because ARPU will vary significantly by user segment and domain pack adoption.

Revenue Expansion: The Domain Pack Multiplier

Each new domain pack creates a revenue multiplier: it attracts new users from that domain AND generates expansion revenue from existing users who add it.

8. Risk Assessment

Technical Risks

Business Risks

Competitive Risks

Domain Pack Risks

9. Success Criteria

Launch Success (First 90 Days)

Growth Success (6-12 Months)

Platform Success (12-24 Months)

Product Success (Engagement)

10. Investment Requirements Framework

Development Investment

Note: Investment depends on build vs. buy decisions, team composition, and scope. Framework provided; specific numbers require planning decisions with engineering.

Infrastructure Investment

Operational Costs

Bootstrap Path

*Based on $162/month infrastructure cost. Exact subscriber count depends on final usage-based pricing and ARPU.

11. Recommendation

Go Decision: PROCEED

Rationale:

Key Assumptions to Validate

Key Decision Points Required

Next Steps

Appendices

A. Domain Pack Expansion Roadmap

B. BYOT Model Precedents

C. Platform vs. Point Solution Economics

D. Infrastructure Cost Reference

Full infrastructure cost breakdown: See Legionis/Product/infrastructure-cost-breakdown.md (validated Feb 13, 2026, $162/month total, detailed per-service analysis).

E. Key Assumptions Register

Document Control

Business case prepared by BizOps | Legionis (legionis.ai)