Method
Marta Warren’s AI visibility method
AI does not discover brands through one channel. It understands them through a network of signals: media coverage, website copy, reviews, third-party profiles, creator content and search results. Marta Warren’s method aligns those signals so AI recognises a brand clearly, trusts what it reads and recommends it with confidence.
The same method sits behind every service I deliver, and agencies put their name to it when they sell AI visibility through the partnership.
The diagnosis
Three truths of a brand
Every brand has three truths: the intended brand, the published brand and the extracted brand. The gap between them explains how AI describes it.
The first truth
The intended brand
What the brand wants to say about itself. Its positioning, its language, the story in the strategy deck.
The second truth
The published brand
What actually exists in public. The website, press coverage, reviews, listings, awards and third-party profiles.
The third truth
The extracted brand
What AI takes from all of that, and repeats when someone asks.
When the intended, published and extracted brand don’t match, AI fills the gaps with what it can verify. That is how a hotel repositioning as modern and relaxed is still described as grand and opulent, and how a brand that is well known by name disappears when guests ask for a recommendation. Most of the gap sits between the published and the extracted brand: the ground PR already manages.
The outcome
Recognised, Trusted, Recommended
i.
Recognised
A recognised brand is easy to understand and consistently described across channels.
- Clear positioning and category language
- Consistent brand descriptions
- Relevant search and AI visibility
- Clear ownership of what the brand wants to be known for
ii.
Trusted
A trusted brand’s claims are supported by credible sources that can be checked and repeated.
- Authoritative media and third-party mentions
- Reviews, ratings, awards and expert validation
- Clear proof points on its own website
- Consistency across profiles, listings and public sources
iii.
Recommended
A recommended brand appears in the environments that shape decisions, shortlists and AI-generated recommendations.
- Presence where audiences ask questions
- Content that answers real decision-stage prompts
- Strong comparison and category relevance
- Clear reasons to choose it over alternatives
In practice
How the method becomes your service
01
Define what the brand wants to be known for.
The first step defines the sentences you would want a journalist, a trusted advisor or an AI assistant to say about the brand. They become the narrative anchor and the basis of the prompt set.
02
Measure the gap.
Branded and discovery prompts across AI platforms show how the brand is described, who AI recommends instead and which sources shape the answers.
03
Name the type of gap.
A facts gap means the proof does not exist yet. A framing gap means the proof exists but is described in the old language. They lead to very different plans.
04
Build the evidence plan.
The evidence plan sets out four things: the desired association, the evidence required, the authority sources and the PR activity to secure them. This is where the agency’s PR team does what it does best.
05
Track and repeat.
The audit prompts become the baseline, re-run quarterly to show clients what has moved.
Want to see how the method would work for your clients?
If your agency is weighing up an AI visibility service, we can start with a short call.