AMEC Launches GEO Principles to Tame AI-Era Measurement

AMEC, the Association for Measurement and Evaluation of Communication, has launched a set of GEO Principles aimed at bringing rigour to AI-led communications measurement. The framework arrives as practitioners grapple with generative engines, the AI systems that increasingly shape how audiences encounter brands, replacing the familiar search results page.

The name signals the shift the industry body is responding to. GEO stands for generative engine optimisation, the practice of trying to influence what AI platforms say about an organisation. Communications teams have spent two decades learning to measure search visibility; AMEC argues that the AI answer layer demands a different discipline, and that measurement standards written for a search-first world no longer cover the ground.

Why does this matter now?

The principles arrive at a moment when communications measurement is under strain from two directions. First, AI tools now produce summaries, answers and recommendations that audiences treat as authoritative, which makes the old metrics of clicks and impressions look thin. Second, vendors have rushed AI-branded measurement products to market faster than standards have developed, leaving buyers with little way to judge what a claim of "AI measurement" actually contains.

AMEC's move follows its established pattern. The association previously produced the Barcelona Principles, first adopted in 2010 and updated since, which remain the most widely cited framework in communications measurement. The new GEO Principles extend that standards work into territory the Barcelona Principles never anticipated.

What does this change for practitioners?

For in-house teams and agency measurement leads, the immediate value is a shared reference point. When a vendor pitches an AI measurement product, or a client demands "GEO reporting", the principles give practitioners something to measure the pitch against. The framework is voluntary, as AMEC standards are, but the Barcelona Principles showed how far a voluntary code can travel when the industry adopts it as common language.

The practical questions the principles address are the ones measurement teams already argue about internally:

• How do you track whether an AI engine represents your organisation accurately?
• Which inputs does a generative engine draw on, and can communications influence them?
• What counts as an outcome, rather than an AI-generated mention, in reporting to the board?

AMEC frames the stakes in terms of rigour. Measurement that cannot survive scrutiny, the association's long-standing argument goes, ends up as decoration on a report rather than evidence for a decision. Applying that argument to AI-led evaluation is the explicit purpose of the launch.

A standard to watch

Skepticism remains warranted on two fronts. Voluntary principles only bind those who choose to be bound, and the vendor market has a commercial interest in loose definitions. Whether the GEO Principles acquire the authority of their Barcelona predecessors will depend on adoption by major networks, in-house teams and, ultimately, the clients who sign the invoices.

For now, the launch gives practitioners a first structured answer to a question that has been circulating since generative engines went mainstream: how do we prove value when the audience no longer sees the page we optimised? AMEC's answer is that measurement discipline, not new dashboards, comes first. The industry will now test whether that discipline holds.