Research

Basis of the work

Frontier Marketing is developed from three sources: consulting practice with organisations facing changes in discovery and intermediation; applied development of AI agents and retrievability work; and reading of published analysis on AI adoption in organisations, including Microsoft’s work on the Frontier Firm, listed on the references page.

It is not presented as empirical research with a formal method. It is a documented framework: a set of definitions, propositions and layers, stated precisely enough to be examined, disagreed with and revised.

Open questions

• How do language models and agents actually select, compare and recommend organisations, and how stable are those selections over time?
• Which observable properties of an organisation's public material most affect whether it is correctly described by an AI system?
• What organisational structures make an agentic ecosystem a durable capability rather than a set of isolated tools?
• How can the compounding of go-to-market capability be measured without reducing it to channel attribution?
• Which components of a value proposition prove most resilient when the intermediation layer of a market changes?

Standards applied to this documentation

• Claims are stated at the level of confidence the evidence supports; conjecture is labelled as conjecture.
• No source is cited unless it exists, has been consulted, and is reachable at the URL listed on the references page.
• No third party is described as endorsing, adopting or validating this framework.
• Unpublished work is never described as published; publication status is listed explicitly.
• Definitions are held stable across pages; when a definition changes, the framework version and last-updated date change with it.

Revision

Each page carries a published date, a last-updated date and the framework version it documents. Substantive revisions increment the framework version; corrections of wording do not.