Does Organization schema raise the GEO entity/authority score?
Hypothesis
Sites with valid Organization schema score higher on the GEO entity/authority sub-score than sites without it.
Method
Segment the corpus by Organization-schema presence; compare the GEO entity sub-score distributions with 95% CIs; report the delta. Pre-registered; null results published.
Dataset
850k+ corpus (metric: missing_organization_schema - currently 52.2% missing).
Metrics
| GEO entity sub-score delta | measure with 95% CI |
Result
CONFIRMED. Entity/authority sub-score +15.83 (95% CI +15.61 to +16.04), n=33,034 with schema vs 69,840 without. Overall GEO +17.77, of which 1.45 points is definitional (8%). The registered metric was NOT schema-independent as assumed: 77 of its 100 points sit on checks a JSON-LD Organization block can help satisfy. Isolating the 23 points it cannot reach gives a LARGER gap (+20.49, CI +20.12 to +20.87) than the 77 it can (+14.43), so the association is not an artefact. 176 of 178 AI-Ready sites publish it vs 5.6% of sites under score 50.
Decision
ANSWERED. Segment methodology (MTH-SEGMENT); minimum 300 sites per cell or marked provisional. Answered 2026-08-19 on the SAVI index panel wave 1 (102,873 domains, scorer geo-145-2026-07, population-weighted).
Lessons learned
- The registered sub-score was assumed schema-independent and was not. Always read the scorer function, not the check names, before segmenting on a markup signal.
- Effect is a marker, not a lever: the tag is worth 1.45 points while carriers lead by 39 and 29 points on categories with no schema check. The advantage nearly vanishes on AI citability (+3.73), the category closest to actual AI visibility.
- A controlled before/after follow-up is the design that would establish causation.
Confidence
HIGH
Evidence
benchmark -metricKey missing_organization_schemaresult - https://searchscore.io/research/organisation-schema-marker-not-lever/