InfiniteRanks resources
Design an AI visibility index with a published methodology
An index needs a recurring measured dataset and a transparent denominator. A small hand-picked prompt set cannot be presented as a market-wide brand ranking.
Define the population and sample
Name the industries, countries, languages, questions and brands included. Explain how the questions were selected and weighted. Record whether they come from licensed search data, customer research or generated templates. Disclose coverage gaps and the rights needed to publish the underlying data.
Preserve collection provenance
Store capture time, requested and returned model, method, retries, eligibility and citation completeness. Version the question cohort and entity rules. A changing sample can create a false trend even when every individual answer remains the same. Keep the raw records needed to reproduce a published aggregate.
Publish formulas and uncertainty
Explain mention rate, citation rate and any weighting. Show the raw denominator and missing observations. Distinguish a selected tracked-brand share from total market share. Avoid false precision when a small number of answers drive the change. Document the refresh cadence and how corrections affect previous releases.
Use your workspace as a bounded alternative
InfiniteRanks workspace evidence can describe your declared cohort and collection conditions. It is not automatically a public market index. Use the metric guide and source-analysis method to write a transparent internal benchmark while a broader index would require a separately established dataset and publication process.
Sources and further reading
Sources explain the underlying standards. They do not endorse InfiniteRanks or guarantee a ranking outcome.