Wardley Mapping Reference
Purpose: Apply Simon Wardley's value-chain and evolution mapping to visualize the strategic landscape. A Wardley Map plots components on two axes — anchored at user needs (value) and charted across evolution (genesis → custom → product → commodity) — to expose strategic choices, inertia, and climatic patterns.
Scope Boundary
- helm
wardley: Simon Wardley value-chain + evolution mapping (this document). - helm
porter(elsewhere): Industry structure analysis. Porter is static; Wardley is evolutionary. - helm
jtbd(elsewhere): Customer-job reframing. Wardley starts with user needs and traces the chain backwards. - helm
blue-ocean(elsewhere): New market creation. Wardley reveals where on the map to create new space. - Atlas (elsewhere): Software architecture. Wardley is strategic; Atlas is structural.
- Magi (elsewhere): Go/No-Go decision. Wardley provides the terrain for Magi's deliberation.
Core Concepts
Map vs Graph
A map has position and direction. A typical "architecture diagram" is a graph — nodes and edges without meaningful position. Wardley Maps add two axes:
- Y-axis (value chain): visible to user at top; supporting components at bottom. Anchored at user need.
- X-axis (evolution): Genesis → Custom → Product (+ rental) → Commodity (+ utility). Direction of evolution is left-to-right.
Value User Need
(visible) ┼─────────────── A (visible component)
│ ╱│
│ ╱ │
│ B │ (custom-built)
│ ╱│ │
│ C │ D│ (products / rental)
Invisible │ ╱ │ │
┼──────────E──F──G── (commodity / utility)
│ │ │
▼ ▼ ▼
Genesis Custom Product Commodity
←──── Evolution ────→Each component's position encodes its evolutionary stage; each dependency encodes the value chain.
Four Evolution Stages
| Stage | Definition | Examples | Characteristics |
|---|---|---|---|
| Genesis | Novel, chaotic, rare, uncertain | Early research, new invention | Wonder, gamble, high cost, high variance |
| Custom-built | Unique per user, bespoke, limited copies | Early product attempts, consulting builds | Emerging understanding, high labor, learning phase |
| Product (+ rental) | Repeatable product / service with multiple vendors | Off-the-shelf software, subscriptions | Feature comparison, competition on differentiation |
| Commodity (+ utility) | Undifferentiated, cost-of-doing-business | Electricity, S3 storage, HTTP | Competition on cost/operational excellence, standardized |
Direction of evolution is one-way: components move right, not left. Attempting to "reinvent" a commodity as custom is usually strategic suicide.
Characteristics by Stage
| Characteristic | Genesis | Custom | Product | Commodity |
|---|---|---|---|---|
| Ubiquity | rare | uncommon | common | ubiquitous |
| Certainty | uncertain | emerging | stable | standard |
| Customer-perceived value | variable | high | moderate | low (assumed) |
| Failure mode | fatal / breakthrough | feature miss | losing race | commodity margin squeeze |
| Appropriate methodology | experimental | agile | lean / continuous improvement | Six Sigma / Ops |
| Appropriate structure | Pioneer | Settler | Town Planner | Town Planner |
| Culture | speculative | value-creation | efficient delivery | extreme efficiency |
Wardley's Pioneer / Settler / Town Planner (PST) organizational model maps which teams own which stage.
Climatic Patterns
The map reveals universal patterns. Use these to interrogate any position.
Core patterns
- Everything evolves: all components drift rightward over time.
- Efficiency enables innovation: commoditizing one layer liberates investment higher up.
- Inertia grows with ubiquity: users and organizations resist commoditization of familiar products.
- Higher-order systems create new sources of worth: once X commoditizes, Y built on top of X becomes a new genesis.
- Past success breeds inertia: the winning custom-build position becomes the chain to break.
- Red Queen effect: standing still = losing ground (competitors evolve).
- No component is static: if you don't migrate, someone will eat your position.
Doctrine (Universal Principles)
Irrespective of context, good strategy applies these:
- Focus on user needs.
- Use a common language (maps).
- Be transparent — share the map.
- Challenge assumptions — is the component really at that stage?
- Remove duplication — multiple teams building the same custom thing.
- Use appropriate methodology per stage.
- Think small (teams, contracts, services).
- Use standards where appropriate — especially for commodities.
- Design for constant evolution — every commodity was once a genesis.
Gameplay / Strategic Moves
Wardley catalogs dozens of "gameplays." Key families:
| Family | Examples | When appropriate |
|---|---|---|
| Accelerator | Open source a component, sponsor competition | Force commoditization of adjacent component |
| Decelerator | FUD, patents, licensing, regulation | Slow competitor evolution |
| Market | Pricing, differentiation, geography | Shape demand |
| Ecosystem | Platform + third-party innovation (ILC: innovate-leverage-commoditize) | Extract signal from the edge |
| Defensive | Exit, retirement, sunset | Recognize when a position is indefensible |
| Attack | Fool's mate (attack an inert competitor) | When competitor has inertia |
ILC loop (classic Wardley pattern)
Innovate internally → Leverage the ecosystem that forms around you → Commoditize the successful piece, freeing the next layer.
AWS did this: launched S3/EC2 → ecosystem built products on top → AWS folded successful abstractions (Lambda, Aurora) back down to commodity.
Workflow
ANCHOR → identify the user need (who + what job)
→ not "customer" (internal user, admin, ops are users too)
CHAIN → list components required to deliver that need
→ include tech + data + people + processes
POSITION → place each component at the right evolutionary stage
→ consider: ubiquity, certainty, comparable-market evidence
DRAW → plot value chain vertically (user at top)
→ plot evolution horizontally
→ draw dependencies (chain edges)
SCAN-CLIMATE→ identify inertia, duplication, stage mismatches
→ spot higher-order-system opportunities
→ note competitor positions on same map
APPLY → apply doctrine (focus, remove duplication, right methodology per stage)
→ select gameplays (accelerator/decelerator/ecosystem/attack)
DECIDE → make strategic choices with the map as evidence
→ assign Pioneer/Settler/Town Planner ownership
→ hand off to Magi for arbitration if trade-offs existCommon Strategic Mistakes (the map makes visible)
| Mistake | Signal on the map | Correction |
|---|---|---|
| Building custom what is commodity | Component in Custom but industry-wide has it in Commodity | Buy/rent, reassign team |
| Managing commodity with Pioneer methodology | Genesis methodology applied to known commodity | Switch to operational excellence (Town Planner) |
| Standardizing too early | Standard applied to Genesis component | Delay; let variants explore |
| Ignoring higher-order opportunity | Commoditized layer sits dormant, no new genesis above | Invest in genesis above the commoditized base |
| Relying on a single vendor for commodity | No second source | Encourage utility / multi-vendor |
| Pioneer team reassigned to Town Planner work | Wrong structure for stage | Rotate people; use PST |
Output Template
## Wardley Map: [Anchor user need]
### Anchor
- **User**: [who]
- **Need**: [what job — not a feature]
### Value Chain (top → bottom = visible → invisible)
1. [Component A] — visible to user
2. [Component B] — supports A
3. [Component C] — supports B
4. ...
### Evolution Positions
| Component | Stage | Evidence |
|-----------|-------|----------|
| A | Custom | [market signal, competitor analysis] |
| B | Product | [multiple vendors exist] |
| C | Commodity | [utility / price competition] |
| ... | ... | ... |
### Map Builder (ASCII or image)[user need]──[A: Custom]──[B: Product]──[C: Commodity] [D: Custom]─────┘
### Climate & Inertia Findings
- [ ] Component X has inertia — [why]
- [ ] Component Y is positioned wrong (we custom-build but industry-commoditized)
- [ ] Duplication: team α and β both build [Z] — consolidate
- [ ] Higher-order opportunity: commoditized [C] enables new genesis [N]
### Strategic Moves
| Move | Component | Rationale |
|------|-----------|-----------|
| Accelerator: open-source | [X] | force [Y] to commodity |
| Ecosystem: ILC | [platform] | leverage third-parties, fold back |
| Defensive: sunset | [Z] | indefensible position |
### Pioneer / Settler / Town Planner
- **Pioneer team** owns: [genesis components]
- **Settler team** owns: [custom → product transition]
- **Town Planner team** owns: [product → commodity, ops]
### Doctrine Check
- [ ] Anchored at user need
- [ ] Map shared with stakeholders
- [ ] Appropriate methodology per stage
- [ ] Duplication removed
- [ ] Standards applied where appropriate
### Handoffs
- Atlas: architecture follow-through on commoditized / standardized layers
- Shift (`detect`/`modernize`/`radar`): modernization plan for components that must evolve (absorbed from horizon)
- Magi: Go/No-Go on strategic moves
- Spark: genesis opportunities for new features
- Ledger: cost implications of commodity vs custom
- Scribe: written strategy documentAnti-Patterns
| Anti-pattern | Fix |
|---|---|
| Drawing a graph, not a map | Enforce Y=value, X=evolution; position must mean something |
| Not anchoring at user need | Start with user + job, not technology |
| Mixing evolution with maturity | Evolution = industry-wide stage, not how mature your implementation is |
| Static map (draw once, forget) | Re-map when competitor moves or new commoditization hits |
| Ignoring inertia | Name it explicitly; plan against it |
| Custom-building commodities | Buy/rent; reserve engineers for genesis |
| Same methodology everywhere | Apply experimental / agile / lean / Six Sigma by stage |
| PST ownership mismatch | Rotate Pioneer/Settler/Town Planner by component stage |
Deliverable Contract
When wardley completes, emit:
- Anchor with user and need.
- Value chain from visible (user-facing) to invisible (supporting).
- Evolution positions for every component with evidence.
- Map builder (ASCII or image).
- Climate & inertia findings.
- Strategic moves chosen with rationale.
- Pioneer / Settler / Town Planner ownership assignments.
- Doctrine checklist.
- Handoffs: Atlas, Shift (
detect/modernize/radar), Magi, Spark, Ledger, Scribe.
References
- Simon Wardley — Wardley Maps (CC-licensed book, Medium series)
- Learn Wardley Mapping — learnwardleymapping.com (tutorial)
- Ben Mosior — Hired Thought blog and YouTube
- onlinewardleymaps.com — free browser-based mapping tool. The June 2025 modernisation release (docs.onlinewardleymaps.com 2025-06-15) ships a TypeScript rewrite, a resizable split-pane editor, and a simplified DSL with new decorator syntax (legacy maps still parse).
- Topographical Intelligence for Your Business — HBR write-up of the framework
- LEF (Leading Edge Forum) — Wardley's original research
- "Crossing the River by Feeling the Stones" — Wardley's essay on the doctrine
- Mark Craddock — Learn Wardley Mapping (Medium, Apr 2026) — Claude Code skill benchmarked blind against 25 of Wardley's own published maps; documents the OWM-DSL-as-LLM-target pattern.
- Mermaid Wardley Map support — community PR landed 2025; LLMs can now emit Wardley diagrams directly into Mermaid blocks for inline rendering.
- Simon Wardley — Craft Conference 2025 (Budapest, May 2025) keynote on mapping the AI value chain; also
Agile Meets Architecture 2025speaker billing.
2025-2026 AI Tooling for Wardley Maps
Three classes of AI tooling now coexist; treat them as Tier 4 defaults, never as truth:
| Class | Examples | Use | Risk |
|---|---|---|---|
| Claude / GPT skills emitting OWM-DSL | haberlah/wardley-mapping (Claude Code skill, 2026), IanSimon23/wardley_maps |
First-draft maps from a scenario description; export to onlinewardleymaps.com | LLM may collapse Custom and Product when the industry signal is ambiguous — always demand an evidence column |
| LLM-in-Mermaid | community Mermaid plugin (2025) | Inline maps in chat or PR comments | No interactive editing; positions are not stable across regenerations |
| Drag-and-drop with AI coach | IanSimon23/wardley_maps, MCP-market "Wardley Strategy Mapping" skill (2026) |
Workshops and reviews | Coach text reflects training-set bias toward AWS / SaaS exemplars; sanity-check on non-software domains |
Always require the AI to attach an Evidence column per component (per the Output Template) and to name the inertia it sees — otherwise the map is graph, not map.