AI use statement

Where we use AI

We use AI as operational infrastructure for tasks including, but not limited to:

• research and competitive analysis

• organising unstructured data

• stress-testing assumptions and personas

• building measurement frameworks

• reviewing our own work for gaps and consistency

• writing support across documents, presentations and content

AI reduces manual effort on the repetitive parts of our work so we can redirect that time to the more productive elements of our consulting work for clients. Nevertheless, every deliverable is reviewed by an expert human, most often our founder, Andrés López-Varela, and reflects our professional judgment and interpretation, whatever tools were used along the way.

Where we don’t use AI

We never use AI to author or create strategic recommendations. The judgment calls in our work all come from our lived experience shaping and running marketing programs and business. Things like what a business should do to grow its customer base, improve its lifetime customer value, the sorts of channels it should be present on for its marketing, which audiences to pursue for greatest return on investment and other mission-critical conclusions are all generated by the brains of our people, not that of a model.

We also don’t use AI to facilitate workshops, design training programs or lead creative development. We don’t use the technology to manage client relationships, which are personal and often include sensitive and confidential information, and are built on the hard-won trust between people. And we don’t use AI to design visual identities, brand systems and graphic or visual content for clients, all of which need a human’s taste, creativity and aesthetic experience to ensure success.

The AI tools we use

We use only commercial-grade tiers of leading AI tools, selected and paid for by us; we don’t accept commercial or in-kind partnerships from any AI or other technology vendor. We choose services and technology with terms that ensure our inputs are not used to train public models or make similar improvements. The specific tools we use change as the market and best-practice evolves, which is partly why this is a declaration and not a fixed list of tech.

Client information

We don’t put confidential client information into consumer or public AI services. We also don’t use client materials or client information to train or fine-tune any AI model, including any private ones we might use. If a client has specific requirements about AI usage, we agree those before work begins.

Our working records

The prompts, working files, notes and AI interaction records we use are part of how Marea works, in the same way that previous generations of consultancies kept working papers. They are confidential information and, in many cases, represent our Intellectual Property, and these stay with us.

Disclosure

We talk openly about our use of AI. It’s in our Master Services Agreement, we discuss it in our marketing content and are happy to answer any questions about how we use the technology. This statement describes our current practices and is updated as they evolve. This statement is informational and doesn’t form part of any agreement, unless expressly incorporated.

This statement was last updated in September 2026.