The 8 Dimensions Behind Your AI Visibility Score

An AI visibility score should show more than whether a brand appeared in an answer. GeoNeo Methodology V3.3.2 measures eight distinct signals — from presence and recommendation to authority and exclusivity — so teams can see not only if they are visible, but how strongly AI engines position them.

The GEO Score is measured out of 100. It is built from 16 queries across eight intent categories and assessed across Western and Asian AI engines. Each dimension answers a different commercial question, which matters because a brand can perform well in one area and remain weak in another.

For example, an organisation might appear frequently but rarely be recommended. Another might be strongly recommended by one engine yet absent from the rest. A single mention count would treat those situations as similar. The eight-dimension framework does not.

1. Presence — 20 points

Presence measures how often a brand appears across the queries and engines included in a scan. It is the most basic visibility question: when buyers ask relevant questions, does the brand enter the answer at all?

Imagine that a business is well known in traditional search but appears in only a small share of AI answers about its category. Its website may still attract traffic, yet it is missing from the shortlists being created inside AI tools. A low Presence score makes that gap visible.

Presence is worth 20 points, but it cannot explain the whole picture. Being named does not mean being endorsed, trusted or placed ahead of competitors. It is the foundation on which the other dimensions build.

2. Recommendation — 20 points

Recommendation measures whether an AI engine actively suggests the brand, rather than merely mentioning it. This distinction is commercially important because a neutral reference and a positive recommendation can affect a buyer very differently.

Consider an answer that lists five providers. A brand included near the end as an alternative is visible, but it has not necessarily won the recommendation. Another answer might identify that brand as the best fit for a particular use case and explain why. Both answers contain a mention; only the second shows strong recommendation quality.

Recommendation carries 20 points, equal to Presence. Together, the two dimensions separate “AI knows the brand” from “AI is prepared to suggest the brand”.

3. Placement Hierarchy — 15 points

Placement Hierarchy measures where a brand appears within an AI response. Brands introduced early or positioned as a leading option receive a stronger signal than brands buried near the end.

This reflects how people consume answers. Many users act on the first clear recommendation rather than reading every sentence or checking every alternative. A brand that appears consistently but only after several competitors may have a visibility problem that a simple mention rate would miss.

Placement Hierarchy is worth 15 points. It helps teams ask a sharper question: are we merely present, or are we prominent?

4. Topic Coverage — 15 points

Topic Coverage measures the breadth of areas in which a brand appears across the eight intent categories: Recommendation, Comparison, Trust, Problem, Use-case, Innovation, Sector and Competitive.

A brand may be visible for branded questions while disappearing from broader category or problem-led questions. That often means the AI system can recognise the name but does not connect it strongly enough with the needs that lead buyers towards a purchase.

For example, a financial-services brand might appear when users ask about the company directly but not when they ask which provider is suitable for a particular customer type. Topic Coverage identifies that narrow association.

The dimension is worth 15 points. GeoNeo also applies an anti-inflation safeguard: when a brand appears in fewer than three categories, the total GEO Score is capped at 65. This prevents a small number of isolated mentions from producing an overstated result.

5. Sentiment — 10 points

Sentiment measures whether the brand is framed positively, neutrally, negatively or with mixed signals. Visibility is not automatically valuable if the surrounding description creates doubt.

An answer might cite a company as a known provider but immediately raise concerns about suitability, service or risk. Another might describe the same company as dependable for a specific requirement. Both answers increase raw visibility, but their likely commercial effect is different.

Sentiment contributes 10 points. It gives marketing and communications teams a way to distinguish between being discussed and being positioned favourably.

6. Consistency — 10 points

Consistency measures whether a brand is cited across multiple engines rather than appearing in only one. This matters because buyers do not all use the same AI platform, and each engine may rely on different retrieval systems, sources and model behaviour.

A brand that performs well in ChatGPT but is absent from Gemini, Perplexity, DeepSeek or Qwen has a fragmented visibility profile. Its result may depend too heavily on one ecosystem. A more consistent brand appears across several relevant engines, even if the wording varies.

Consistency is worth 10 points. For organisations operating across Asian and Western markets, it is especially useful because it prevents strength in one group of engines from hiding weakness in another.

7. Authority — 7 points

Authority measures citation strength with additional weight for Tier-1 engine citation. The aim is to recognise that not every appearance carries the same level of influence within the framework.

Authority is not a generic claim that one piece of schema or one technical change can make a brand trusted by AI. It is a measured output from the scan. Teams should therefore treat it as a diagnostic signal, then investigate the content, sources and entity information associated with stronger or weaker results.

At seven points, Authority is smaller than Presence or Recommendation but still material. It adds quality context to the broader visibility picture.

8. Exclusivity — 3 points

Exclusivity measures whether a brand is mentioned alone rather than being co-cited with competitors. It is the smallest dimension at three points, but it can reveal a powerful competitive signal.

If an AI answer repeatedly names a brand alongside four rivals, that brand is part of the consideration set. If it names the brand alone as the clear answer to a relevant question, the positioning is stronger and more distinctive.

Exclusivity should not be interpreted as a promise that a brand can eliminate competitors from AI answers. Instead, it shows how often the brand earns an unshared position within the measured responses.

How the dimensions work together

The eight dimensions should be read as a profile, not eight isolated grades. Presence and Recommendation carry the most weight, but Topic Coverage explains whether visibility is broad. Placement shows prominence. Sentiment shows framing. Consistency checks cross-engine strength. Authority adds citation quality, while Exclusivity captures unshared positioning.

GeoNeo's compare mode also separates branded and organic behaviour. The final compare blend is organic × 0.55 plus direct × 0.45, while competitors are scored on organic queries only. This creates a more neutral comparison than testing competitors with questions that already contain their names.

The result is a scoreboard for diagnosis and improvement. It does not guarantee future recommendations, and it should not be reduced to a single vanity number. The useful question is what caused the score — and which weak dimension represents the most important commercial gap.

For the complete scoring framework, read the GeoNeo methodology.

Turn the score into a next step

A GEO Score gives teams a baseline across the signals that shape AI visibility. GeoNeo then uses a managed six-step workflow — Brand Profile, GEO Score, Visibility Probe, Gap Assessment, Gap Content and Prompt Watch — to move from measurement towards action.

Run your free scan to see how AI engines currently mention, position and recommend your brand.

Frequently asked questions

What is the GeoNeo GEO Score?

The GeoNeo GEO Score is a 0–100 measure of brand visibility across eight dimensions: Presence, Recommendation, Placement Hierarchy, Topic Coverage, Sentiment, Consistency, Authority and Exclusivity. It follows GeoNeo Methodology V3.3.2.

Why are Presence and Recommendation separate?

Presence measures whether a brand appears, while Recommendation measures whether an AI engine actively suggests it. A brand can be frequently mentioned without being positively recommended, so combining the two would hide an important commercial difference.

Why is the score capped when coverage is narrow?

If a brand appears in fewer than three intent categories, GeoNeo caps the total score at 65. This anti-inflation safeguard prevents a few isolated mentions from creating an exaggerated impression of broad AI visibility.

Does a high GEO Score guarantee AI recommendations?

No. AI answers vary by engine, prompt, market, available sources and update cycle. The score is a structured measurement of observed visibility, not a guarantee of future ranking or recommendation.

How many queries are used in a full scan?

GeoNeo uses 16 queries across eight intent categories: Recommendation, Comparison, Trust, Problem, Use-case, Innovation, Sector and Competitive. Each category includes one organic query and one branded query.