Skip to content
Get my AI grade

Published method

Public Cloud Platforms answer source visibility

Ranks URL-derived citation hostnames in the complete public prompt benchmark without additional provider calls.

2-parent-f0e56a0391a3694a

Version

Jul 27, 2026

Published

1

Board

cloud-platforms-source-visibility

Method ID

Sources

Where the facts come from

Provider Count

3

Provider Matrix Version

64232fd32583

Source Methodology Version

2-parent-f0e56a0391a3694a

Ranking rule

How the order is produced

unique cited-answer coverage descending; engine breadth descending; median answer-level citation position ascending; mean answer-level citation position ascending; hostname ascending

Refresh

When the method reruns

  • The prompt boards are re-run in rotation, about one category a day, so each board refreshes every few weeks and shows the date of its snapshot.

Boards

Boards using this version

Every published board built with this method. Each opens on its latest snapshot.
BoardPrimary measureRowsConfidenceSnapshotBoard versionRefreshBoard
Most cited domains for cloud buying promptspublic-prompt-source-answer-coverage128100%Sep 21, 20262-parent-f0e56a0391a3694aRotating scheduleOpen the board: Most cited domains for cloud buying prompts

Versioned contract

Complete measurement ledger

These fields are stored with the methodology version. A changed contract must publish a new version before comparable movement can resume.

Scope

cloud-platforms

Cohort

Not specified.

Semantics

This is sourcing visibility in structured model/API citations. A citation is not a recommendation, endorsement, authority score, quality judgment, correctness judgment, or consumer-app share.

Limitations

  • The board covers one fixed Cloud Platforms buying-prompt category, not all topics or all AI answer surfaces.
  • Models can vary between runs; the board is one complete daily cohort across the configured model/API versions.
  • Citation position is the provider adapter's normalized structured-citation ordering, not a prose sentiment measure.

Prompt Count

16

Rank Formula

unique cited-answer coverage descending; engine breadth descending; median answer-level citation position ascending; mean answer-level citation position ascending; hostname ascending

Corpus Version

cloud-platforms-2026-07-27

Provider Count

3

Evidence Policy

Source Evidence

validated canonical citation URL, URL-derived hostname, position, provider artifact hash, and aggregate artifact hash

Provider Calls For This Board

0

Full Provider Answers Published

No

Full Provider Answers Stored In Public Index

No

Benchmark Version

4-cloud-platforms-2026-07-27-models-64232fd32583

Canonicalization

  • Accept only absolute HTTP(S) URLs with public-looking DNS names.
  • Derive the hostname from the URL; ignore provider domain labels.
  • Remove fragments and normalize default HTTP(S) ports.
  • Deduplicate each canonical URL within one prompt/provider answer.
  • For duplicates, keep the best positive 1-based position.

Metric Denominators

Mean Position

sum of validated answer-level citation positions divided by the exact number of deduplicated answer-URL observations

Engine Breadth

3 providers

Answer Coverage

48 successful prompt/provider answers

Median Position

middle validated answer-level citation position after sorting; for even samples, the midpoint of the two central positions

Provider Matrix Version

64232fd32583

Source Methodology Version

2-parent-f0e56a0391a3694a

Required Successful Answers

48

See how often AI names you.

Enter your website. In about two minutes, Rank.ai asks ChatGPT, Claude and Gemini 12 questions your customers ask and grades how often they name you.

  • Your grade out of 100How often AI names you, cites your site, and how high it ranks you.
  • Who gets namedEvery competitor in the answers, most named first.
  • The pages AI readsThe sources behind each answer.
  • Three fixesWhat to fix first, with a brief for the first article.