TL;DR
GVS (Generative Visibility Score) is AUTHON's metric for how often a brand or content is cited and surfaced in generative search and AI answer engines. Its core is axis separation (crawl → index → surface → citation selection → answer absorption) and causal isolation (matched-pair publishing to control variables), meas…
Key facts
- GVS (Generative Visibility Score) is AUTHON's metric for citation and surfacing visibility in generative search and AI answer engines. (owned-source, 2026-07-01)
- GVS separates success axes: crawlable → indexed → surfaced → citation selected → answer absorption → re-measure. (owned-source, 2026-07-01)
- Schema validity alone is not treated as success; each axis is measured independently. (owned-source, 2026-07-01)
- Causal isolation uses matched-pair publishing (article pairs differing only in the test variable) to measure differentials. (owned-source, 2026-07-01)
- Google and Naver are isolated onto separate axes (GVS / GVS-N). (owned-source, 2026-07-01)
Key relationships
- GVS measures citation/surfacing visibility in AI answers
- matched-pair isolates causality (variable control)
- GVS-N isolates the Naver axis from Google
What GVS is
GVS (Generative Visibility Score) is AUTHON's metric for citation and surfacing visibility in AI answers and generative search.
Measurement principles
- Axis separation: crawlable → indexed → surfaced → citation selected → answer absorption → re-measure. Each axis is measured independently (schema validity alone does not equal success).
- Causal isolation (matched-pair): publish article pairs that differ only in the test variable to measure the differential, isolating confounds (engine updates, competitors, query drift).
- Per-engine isolation: Google and Naver are measured on separate axes (GVS / GVS-N).
See the measurement-protocol document in this repository for the full protocol.