Illustrative case · Not a client result
How we turn an answer gap into a testable work queue.
This example explains the analysis structure. It does not contain fabricated client names, results, or performance claims.
1. Hold the question set steady
Start with buyer-intent prompts that define a real evaluation task. Record platform, market, date, interface, answer, mentioned brands, recommendation language, and citations.
2. Separate absence from evidence
A competitor mention alone does not explain causality. We compare product clarity, page usefulness, entity consistency, cited-source eligibility, and third-party corroboration without claiming any single factor caused the answer.
3. Prioritize work by usefulness
The resulting queue may include a stronger comparison page, clearer product entity facts, schema corrections, source outreach, or measurement changes. Each action has an owner and a reason.
Check your visibility gap