Public measurement contract

How AICloudStrategist measures search and AI visibility

A fixed, provider-specific benchmark with evidence, limitations and a declared review cadence.

Version 1 · 6 October 2026

A fixed benchmark prevents selective success claims.

The commercial benchmark contains 32 unbranded problem prompts across India and selected international markets. Every prompt maps to one of the two canonical offers. Results are recorded separately for each provider because search results and AI answers differ by market, account, model and date.

Search evidence

Record provider, locale, device, date, query, observed position and result URL. Top five means positions one through five in the inspected organic result set.

AI evidence

Record product, model, account state, prompt, answer date, recommendation order and cited URLs. A crawler allow rule is not a recommendation.

Business evidence

Join landing page, source/referrer, CTA and accepted lead. Qualification and revenue remain separate owner-maintained stages.

Measurement cadence

  • Weekly: technical eligibility, sitemap, priority-route and event-collector health.
  • Every 28 days: unchanged Google, Bing and available AI prompt set.
  • Every 90 days: review service-market fit, intent ownership and evidence quality before changing the benchmark.

Download the machine-readable benchmark · Download the pre-release baseline.