Our Methodology

The AI Perception
Audit™

An independent diagnostic built on reproducible protocols, executive-grade reporting, and zero conflict of interest. Not a content audit. Not a software dashboard. A strategic diagnostic.

The Framework

Five dimensions.
One clear picture.

1
LLM PresenceMention Score
Frequency and regularity with which the organisation is mentioned in AI responses for strategic queries related to its sector, services and market. A low presence score indicates structural invisibility — the most fundamental form of AI risk.
2
AccuracyAccuracy Score
Degree of alignment between what AI systems say about the organisation and current operational reality. Services, geography, positioning, leadership: every inaccuracy constitutes a reputational and commercial risk — stated by AI with the same confidence as accurate facts.
3
PositioningCompetitive Rank
The organisation's rank in spontaneous AI recommendations against its direct competitors. On a query such as "best [sector] for [need]", is the organisation cited first, third, or not at all? The hierarchy established by AI directly influences conversion rates.
4
International VisibilityCross-lingual Score
Consistency of AI perception across languages and geographic markets. Organisations that produce content primarily in one language often discover their visibility collapses on English-language queries — silently losing opportunities on expansion markets.
5
SentimentTrust Signal
Tone of AI responses associated with the organisation: positive, neutral, or carrying negative signals. A neutral sentiment can be as damaging as a negative one if competitors benefit from clearly positive positioning in the same responses.

Scoring methodology

The VisibilityScan™ Scale

Each organisation receives a composite score from 0 to 100, calculated across all five dimensions and seven AI platforms.

0 – 25InvisibleNo mentions or erroneous mentions across AI platforms.
26 – 45InsufficientSporadic mentions, often inaccurate or incomplete.
46 – 65AdequateRegular presence but not optimised for AI citation.
66 – 80Good VisibilityFrequent and generally accurate mentions across platforms.
81 – 100AI AuthoritySystematic reference, content cited as authoritative source.

Proprietary Framework

VisibilityScan™

VisibilityScan™ is our proprietary analysis framework and scoring protocol, developed in-house over multiple audit cycles to ensure reproducible, benchmarkable and decision-grade results.

Every query, scenario and scoring criterion is documented and standardised. Your February 2026 result is directly comparable with your August 2026 result. You can measure progress.

Unlike automated SaaS tools, VisibilityScan™ is executed by human analysts who interpret results in strategic context — not just surface metrics.

15
Standardised Research Scenarios 4 angles: Identity & Reputation, Recommendation & Comparison, Trust & Risk, International Visibility. Each tested in the target language and in English.
6
AI Platforms Audited ChatGPT, Claude, Gemini, Perplexity, Copilot, Grok and Meta AI — each treated according to its specific behavioural and sourcing characteristics.
/100
AI Perception Index Score Normalised score across 5 dimensions, enabling sector benchmarking and progress tracking over time.

Our Services

Three levels of engagement.
One standard of quality.

From the express Pre-Report to the Full AI Perception Audit™ and continuous Quarterly Monitoring — every diagnostic is independent, confidential, and built for executives.

View pricing →