How AI representation is actually measured.
The SEOWEBI Visibility Pyramid™ defines six pillars of AI representation. The SEOWEBI Visibility Score™ measures your company against all six, directly and with evidence.
Built for position, blind to inclusion.
For two decades, visibility was measured through rankings, traffic, and impressions. That model assumed users would click, compare, and decide. AI systems changed that model structurally. They generate answers. And those answers include only a small number of companies.
In AI-generated answers, visibility is not only about position. It begins with inclusion.
Most measurement systems were built for position. They have no column for inclusion. If your company is not included in the answer, it may never enter that stage of the buyer's decision.
Representation exists before discovery.
Most analytics dashboards still report what happens after discovery - traffic, clicks, conversions. AI representation exists before that. It is the layer that determines whether discovery happens at all.
It is determined by whether AI systems recognise your company as a known entity, trust your information as accurate and authoritative, and select your company as part of an answer. None of those three conditions appear in a standard analytics stack. No column. No alert. No trend line.
What you measure today
Traffic · Rankings · CTR · Conversions
What determines AI representation
Entity recognition · Citation status · Extraction quality · Share of voice in answers
Six pillars of AI representation.
Whether AI systems can reach and process your infrastructure. If they cannot access your content reliably, every other pillar becomes irrelevant.
Whether AI systems can extract meaning from your content. Extraction is different from ranking.
Whether AI systems can identify your company as a distinct entity. Ambiguity reduces citation.
Whether AI systems trust your content as a credible, experienced source. Authority is evidenced, not claimed.
Whether AI systems actually cite, recommend, and correctly represent your company. A company can rank well and still be uncitable.
Whether AI systems have sufficient evidence to recommend your company - whether it is worth recommending.
These six pillars work together. Strength in one does not compensate for absence in another.
A dependency system, not a list.
The six pillars are not independent. Each pillar creates the conditions for the pillar above it to function. When a pillar is weak, the signal chain breaks - not just at that pillar, but at every pillar above it that depended on it.
- Strong performance without entity clarity → AI can access your content but cannot identify who it belongs to.
- Clear entity without topical authority → AI recognises your company but does not trust it as a source.
- High authority without citation readiness → AI trusts your content but cannot extract from it cleanly.
- Citation-ready content without brand demand → AI can cite your company but has no external signal to prioritise it.
AI systems do not evaluate one signal. They reconstruct your company from all available signals simultaneously.
AI systems do not rely only on your website.
They reconstruct your company from accumulated signals: structured data in your code, content across your pages, external references from other sources, consistency of signals across the web, and patterns learned during training.
Your website is one input. It is not the only one. It is not always the most important one. What you believe your company is - and what AI systems have reconstructed your company to be - can be significantly different. That gap is invisible without measurement.
The reconstruction is already happening. The only question is whether it is accurate.
Three recurring failure patterns.
Companies rarely fail in only one obvious place. Naming the pattern is the first step toward measuring it.
Absent from comparison answers. When someone asks an AI system to compare companies in your category, your company is not included. You are not ranked lower. You are simply not there.
Excluded from the citation loop. Established companies dominate AI citation patterns through accumulated reinforcement. The more they are cited, the more they are cited.
Invisible despite existing content. The content exists and is indexed, but AI systems cannot extract clear, citable answers from it. A content architecture problem, not an authority problem.
These patterns are measurable. Once identified, they can be addressed directly.
Observed representation, separated from supporting signals.
Level 5 measures what AI systems returned during the test: presence, position, representation quality, answer specificity, and recommendation sentiment.
Levels 1–4 and 6 assess the supporting conditions that may explain those outcomes: performance, structure, entity clarity, authority, and market demand.
Both are shown separately. A strong observed AI presence is never hidden by missing infrastructure data, and an infrastructure score is never presented as proof of what an AI engine said.
The structure and the measurement.
The Visibility Pyramid™ is the structure. The SEOWEBI Visibility Score™ is the measurement. The AI Brand Intelligence Report™ applies this methodology directly to your company. A minimum of 30 structured prompts, expanded for your named competitors, are run across four AI engines. The results are combined with the six-pillar analysis.
The output is a directly observed AI result, a six-pillar breakdown, an overall score with a confidence label, and a named failure pattern. Not theory. Evidence.
Measure where your company stands.
Understanding the methodology is the first step. Measuring where your company sits across all six pillars is what creates clarity. The AI Brand Intelligence Report™ tells you exactly where you stand - not in theory, but against the AI engines buyers increasingly use to compare companies.