Direct answer
Create prompts that represent relevant ‘best tools’ decisions, then review Monitor > Response Analysis and Cockpit > Compare. Record whether SELF appears, which DIRECT products appear, how each is described, and which source URLs are cited.Data required
Use a stable set of configured prompts, provider and model, period, topic or tag, stored answers, SELF/DIRECT classifications, mention occurrences, ordering where present, sentiment or claims, and citation URLs.Workflow
1
Define relevant prompts
Choose audience, use case, constraints, and market context. Avoid broad list prompts that do not match your product.
2
Review answer-level presence
Read each stored answer and record SELF and DIRECT occurrences separately from rank or recommendation language.
3
Compare narratives and sources
Identify recurring selection criteria, product claims, omissions, and citation URLs. Verify factual claims against approved sources.
4
Monitor the stable cohort
Repeat the same prompt and provider/model cohort and report distributions rather than a single answer.
Interpretation
Product Share of Voice is SELF mention occurrences divided by SELF plus DIRECT occurrences. It does not measure list position, preference, or conversion. Inclusion in a stored answer is one observation and does not guarantee future recommendation.Possible next actions
- If SELF is absent from relevant prompts, test content that provides verifiable evidence for the observed selection criteria.
- If a recurring claim is inaccurate, follow the misinformation correction workflow.
- If one DIRECT product dominates occurrences, inspect its cited sources before choosing a distribution hypothesis.
Limitations
Recommendation lists can change between runs and depend on prompt wording, model, provider, web access, and location. Qwairy observes selected stored answers and cannot infer purchasing behavior or guarantee inclusion.Related pages
Answer Details
Report on answer-level recommendation data.
Analyzing Answers
Review stored answers in context.
Qwairy MCP
Use supported MCP tools and schemas.
Consideration-stage presence
Analyze brand presence during consideration.
B2B evaluation positioning
Investigate evaluation narratives.
Share of Voice
Apply the product occurrence formula.

