Framework
The Frontier Marketing Framework
The documented implementation of Frontier Marketing consists of seven connected layers.
Frontier Marketing combines resilient strategy, resilient value propositions, entity design, retrievability, agentic ecosystems, compounding go-to-market systems and continuous frontier exploration.
The layers are ordered from the most durable to the most frequently replaced. Each layer has a definition, a purpose, inputs, outputs, a relationship to the other layers, and a practical example. The framework is readable as text; the visual representation on the homepage is an illustration of this list, not a substitute for it.
The seven layers
Layer 1
Strategy
Frontier Marketing defines strategy as the durable choice of where an organisation competes, for whom, and why it should remain relevant as the market changes.
Purpose. To provide a stable layer that survives the replacement of tactics, tools and channels.
Inputs
- Market structure analysis
- Buyer and problem definition
- Competitive and substitution analysis
- Organisational capability and constraints
Outputs
- A stated strategic direction
- Explicit boundaries of where the organisation does not compete
- Criteria against which every other layer is evaluated
Relationship to other layers. Strategy governs all six layers below it. No layer may be adopted because it is available; each is adopted because it serves the stated direction.
Example. A B2B infrastructure vendor decides it competes on migration risk reduction for regulated industries, and rejects a broad developer-tooling narrative even though that narrative generates more traffic.
Layer 2
Resilient Value Proposition
A resilient value proposition is a differentiator that remains true and defensible as buyers, problems and technologies evolve.
Purpose. To ensure the organisation remains non-substitutable when comparison is performed quickly and at scale, including by machines.
Inputs
- Strategic direction
- Evidence of a specific buyer pain
- Proof of a distinctive capability
- Scenarios describing how the buyer's problem may change
Outputs
- A stated differentiator with supporting evidence
- A resilience assessment of that differentiator
- A plan for what must become true for it to hold
Relationship to other layers. Receives direction from strategy; supplies the claims that the entity layer must express and that retrievability must make legible.
Example. A logistics software company replaces the claim 'the most complete platform' with a specific, evidenced claim about customs exception handling, then tests whether that specificity survives the arrival of automated brokerage agents.
Layer 3
Entities
Entities are the organisations, people, products, services, problems, categories, concepts and technologies whose explicit relationships define meaning in a market.
Purpose. To make the organisation's meaning explicit and machine-interpretable rather than implied.
Inputs
- Value proposition claims
- Inventory of the organisation's people, products, categories and concepts
- The relationships that connect them
Outputs
- An entity map with stable identifiers
- Structured data expressing entities and relationships
- Consistent naming across all public surfaces
Relationship to other layers. Turns proposition into structure; supplies the substrate that retrievability and agentic systems consume.
Example. A consultancy gives each concept it uses a single canonical page with a stable identifier, and links people, publications and services to it explicitly instead of mentioning them in prose alone.
Layer 4
Retrievability
Retrievability is an organisation's capacity to be found, correctly understood, distinguished and recommended by humans, search systems, language models and autonomous agents.
Purpose. To ensure that what the organisation means is what retrieval systems reconstruct.
Inputs
- Entity architecture
- Content that states claims in self-contained form
- Technical crawlability and server-rendered HTML
- External citations and corroboration
Outputs
- Findability, interpretability, distinguishability and recommendability
- Canonical definitions that can be quoted intact
- Monitoring of how systems describe the organisation
Relationship to other layers. Depends on entities; feeds the agentic layer, which cannot act well on an organisation it cannot interpret.
Example. A company publishes its own definition of its category as a self-contained sentence in server-rendered HTML, so that a model summarising the category can quote it without reconstruction.
Layer 5
Agentic Ecosystems
An agentic ecosystem is a connected set of AI agents, data and workflows that together form organisational capability rather than isolated experiments.
Purpose. To increase what the organisation can perceive, decide and execute without fragmenting into disconnected tools.
Inputs
- Strategic direction and permitted actions
- Access to first-party data and systems
- Definitions of the entities the agents reason about
- Human review and escalation rules
Outputs
- Agents that share context and outputs
- Documented tasks, boundaries and oversight
- Capability that persists beyond any single tool
Relationship to other layers. Consumes entities and retrievability; produces signals and assets that the compounding layer recirculates.
Example. An analysis agent, a content agent and a reporting agent share the same entity definitions and the same first-party data, so an insight found by one becomes usable by the others.
Layer 6
GTM Compounding
GTM Compounding is the design of go-to-market components so that they exchange data, learning, audiences, signals, assets and capability, accumulating capacity over time.
Purpose. To convert marketing activity into accumulated capability rather than isolated campaign output.
Inputs
- Campaign, product, sales and service components
- Shared measurement and identifiers
- A place for learning to be stored and reused
Outputs
- Components whose value increases with the number of cycles run
- Reusable audiences, assets and models
- A documented record of what has been learned
Relationship to other layers. Sits above the operational layers and beneath strategy; any individual component may be replaced without losing the accumulated capability.
Example. A demand programme feeds qualified-conversation transcripts back into proposition testing and into the entity layer, so each cycle sharpens the claims the next cycle uses.
Layer 7
Continuous Frontier Exploration
Continuous Frontier Exploration is the systematic detection, evaluation, testing and adoption of emerging capabilities that are relevant to the organisation's strategy.
Purpose. To keep the organisation at the edge of what is useful without destabilising its direction.
Inputs
- Signals from the market, technology and buyers
- Strategic relevance criteria
- A budget and cadence for evaluation
Outputs
- A reviewed pipeline of candidate capabilities
- Evidence from bounded tests
- Adoption, rejection or deferral decisions with reasons
Relationship to other layers. Feeds every other layer with new capabilities and returns every experiment to strategy for judgement.
Example. A team evaluates a new agent protocol against a stated strategic criterion, runs a bounded four-week test, and records why it was adopted or rejected.
How the layers connect
Strategy sets the criteria. The value proposition states what makes the organisation non-substitutable. Entities express that statement as structure. Retrievability makes the structure legible to humans and machines. Agentic ecosystems act on it. GTM Compounding accumulates the results. Continuous Frontier Exploration returns new capabilities to the top of the stack, where strategy judges them.
Published: 2026-09-20
Last updated: 2026-09-20
Framework version: 1.0
Markdown version: /framework/index.md