Numen Games · Strategy

Wardley Map

A strategic map of the Narrative Work OS. Two axes tell the full story: how visible each component is to the user, and how evolved it is in the market.

Y axis — Visibility

From infrastructure (invisible to users) to visible (what users interact with directly).

X axis — Evolution

From Genesis (novel, unstable) → Custom → Product → Commodity (standardized, available everywhere).

Layer 1

The Foundation

Already commodity. Available everywhere.

Git and LLMs are infrastructure. Interchangeable, cost-optimizable, no strategic lock-in. Every competitor has access to the same foundation.

Layer 2

The Structure

File-over-App. Converging toward standard.

Markdown files in git. No proprietary formats. If every tool disappears tomorrow, the knowledge survives. This principle is the lock-in that works for the user, not against them.

Layer 3

The Differentiators

Custom-built. Nowhere else.

Mission System, Blueprints, Decision Registry, Digital Agents. These components exist nowhere else in this combination. The Mission System is the real moat — it turns work into knowledge artifacts.

Layer 4

The Frontier

Genesis zone. Highest risk. Highest reward.

Agent Identity, Rituals, Narrative Layer. The most differentiating — and the most fragile. Narrative is optional (Layer 4). Rituals require cultural buy-in. Agent Identity is the shortest competitive window.

Full Map Click any node for detail
Visible to user ↑
Genesis Custom Product Commodity
Genesis Custom Product Commodity Narrative Layer Rituals Mission System Digital Agents (CAO) Blueprints Decision Registry Agent Identity Protocols Operational Reports File Layer (MD + Git) LLMs / APIs GitHub / Git
Strategic Tensions
⚠ High

Narrative first = adoption barrier

The most visible component is also the most Genesis. Users hit the lore before the value. The /nwos page is the mitigation — substance first, narrative optional.

⚠ High

LLM commoditization window: 18–24 months

Notion AI, Confluence AI, and Microsoft Copilot will offer native agents with the same LLMs. Using AI is not a moat — it's table stakes. The real moat is what the system doesn't lose: every mission closed with an Execution Reality section, every decision with its rejected alternatives, every blueprint with its delta table. That knowledge can't live anywhere else. An organization with 1 year of NWOS has 365 days of structured institutional memory that can't be imported into Notion. That inertia grows with time.

△ Medium

File Layer: commodity for devs, genesis for orgs

Markdown + git is standard for developers. For traditional organizations, it's an alien workflow. This is why startup ICPs succeed and PYME traditional fails at 10% rate.

△ Medium

Rituals are fragile at scale

Dark Council and Lunar Coven generate the highest cultural engagement. They also generate the highest churn risk if vocabulary isn't adopted quickly. Must be opt-in.

The strategic bet

The moat is what the system doesn't lose

Notion AI can answer questions about what you have in Notion. But it can't answer:

"Why did we make this decision 6 months ago, what alternatives did we reject, and what did we learn executing it?"

→ Notion has documents. NWOS has a Decision Registry — append-only, with rejected alternatives and explicit reasoning.

"What have we learned across the last 20 missions that should change how we approach the next ones?"

→ Notion has tasks with checkboxes. NWOS missions have Epistemic Value and Execution Reality — the learning is the artifact.

"What's the gap between where we are today and where we need to be, and which missions close it?"

→ Notion has pages. NWOS Blueprints have delta tables that map gaps to missions explicitly.

Notion stores what you did. The NWOS stores what you learned doing it. That difference can't be bridged with an AI plugin — it requires redesigning how the organization works from the root.

An organization with 1 year of NWOS has 365 days of decisions with context, missions with Execution Reality, and blueprints updated week by week. That archive is irreplaceable. And it can't be imported into any other tool.

The window: 18–24 months

The goal is to establish Mission System + Decision Registry + Blueprints as the standard for AI-integrated organizations before large players commoditize individual components. Not to be the first to use LLMs — to be the first to build knowledge infrastructure that compounds.

Wardley Map v0.1.0 · Nimrod 🗡️ · 2026-04-07