Prioritizing AI-Search Prompts from Search Volume and Citations

Definition

An opportunity-ranking workflow that estimates demand for buyer prompts from conventional search data, runs those prompts across answer engines and maps which cited URLs recur.

Perspectives

Matt Espinoza (2026-09-13, YouTube)

Generate a large prompt set for the product category, obtain the corresponding Google-query volumes and use them as a rough demand proxy because direct prompt-volume data is unavailable. Run the prompts across ChatGPT, Gemini and Google AI surfaces, then rank opportunities by estimated demand and by the cited pages shared across prompts or engines.

Matt Espinoza described testing 2,000 prompts and estimating AI-search volume at 30% of Google volume. He reported that one Reddit URL could appear across several prompts and engines, but the source provides no prompt list, traffic logs or validation for the 30% multiplier.

How to apply

• Use the output as a dated opportunity queue, not measured prompt volume; validate target engines and question wording through Measuring AI Search Visibility as Crawl, Citation and Click.
• Choose a cited-surface tactic only after the scan: transparent community participation belongs in Earning AI-Search Mentions Through Reddit Participation, while publisher outreach belongs in Pitching Listicles Chosen by Each AI Search Engine.
• Keep Google demand, answer-engine citation, referral click and conversion as separate fields.
• Rescan because citation sources and query wording effects change by engine and date.

Limits

• The 2,000-prompt scale, 30% proxy and 150-million-view claim are vendor self-reports without reproducible exports or independent checks.
• Google search volume may not preserve prompt wording, conversational intent or engine-specific demand.
• A recurring citation identifies a surface already selected by an engine; it does not justify deceptive placement, sockpuppet accounts or manipulation of community sentiment.

Original video