How Viali observes change without overstating certainty.

What Viali means by AI visibility

AI visibility describes whether, where and how a brand appears in answers to relevant customer questions. Useful observations include the brand mention, position or prominence, sentiment, cited sources, competing brands and the complete answer context. A single score is a summary, not the evidence itself.

How tracked prompts are selected

Prompt sets should reflect real market topics, customer personas, buying stages and questions. Viali can connect customer input, category language, search demand and discovered question patterns. Teams retain control over the prompts that represent their market. A prompt set is a measurement instrument and can introduce bias if it is too narrow or unrepresentative.

Why AI answers vary

Answers can differ by model version, platform interface, locale, time, session context, personalization and retrieval path. Viali records the platform, collection time and available context for an observation. Repeated monitoring helps reveal patterns, but it cannot make a probabilistic answer deterministic.

How mentions and citations are identified

Brand mentions are matched against approved brand names and relevant variants, with context retained for review. Citations are captured from links or source references exposed by the answer interface. A brand mention and an owned-site citation are distinct signals: a brand may be recommended without its website being cited.

How market topics are constructed

Related prompts can be grouped by shared intent, audience and buying question. Topic grouping helps teams avoid optimizing isolated phrases and instead understand the broader demand pattern. Human review remains important where a prompt could reasonably belong to more than one topic.

How opportunities are interpreted

Viali connects absence or competitive loss with available evidence: demand, cited authorities, competitor content, owned pages and technical readiness. Priority can reflect confidence, expected impact, effort and business context. Viali does not disclose proprietary weighting and does not present an opportunity score as a guaranteed result.

How website citability is evaluated

Evaluation can include crawl access, robots directives, structured data, entity clarity, direct answers, evidence, authorship, update signals, internal structure and whether important claims are supported. Citability is not a certification and does not mean an answer engine will select the page.

How intervention baselines and outcomes work

1. Record the relevant prompts, engines, citations, owned pages and business signals before implementation.
2. Attach the diagnosis, confidence, intended action, owner and expected signal.
3. Record approval, publication or implementation date.
4. Wait for an appropriate evaluation window.
5. Compare the new observations with the baseline and classify the result as verified improvement, no measurable change or inconclusive.

Attribution limitations

Some AI assistants remove referrer information, answers may change independently of an intervention, and multiple marketing activities may overlap. GA4, Search Console, crawler activity, citations and conversions can strengthen the evidence, but they do not always establish causation. Viali communicates confidence rather than assigning credit automatically.

Locale, personalization and model updates

A result collected in one region or context may not represent every customer. Model and interface updates can alter answers or source presentation without notice. Teams should compare like with like and interpret abrupt changes alongside platform updates.

Data freshness and ranking guarantees

Each observation should display its collection or update time. Older evidence should not be presented as current. Viali does not guarantee rankings, citations, traffic or conversions; it helps teams improve the quality of their evidence, decisions and measurement.