Category: Data & Experiments
Controlled experiments measuring how specific content variables affect AI citation rates. Each experiment isolates a single variable — structure, entity density, freshness, or platform behaviour — and tests it under documented conditions with replicable methodology. Results include raw citation data, statistical significance assessments, and practical recommendations based on observed effect sizes.
-
Fan-out Query Length and Citation Rate: 225 Queries, Inverted Result
Read more →: Fan-out Query Length and Citation Rate: 225 Queries, Inverted ResultPre-registered prediction falsified. 225 Perplexity measurements: short queries cited at 61%, long queries at 16%. The 45pp gap holds across five days. Mechanism: namespace drift, not content quality.
-
Perplexity Cites the Same Pages Every Day: A 14-Day Zero-Variance Replication
Read more →: Perplexity Cites the Same Pages Every Day: A 14-Day Zero-Variance Replication14-day Perplexity measurement confirms zero inter-day citation variance. Q02/Q05/Q06 stable across all runs. T2 rate 0/70. T1 grew from 60% to 82% average days 6–14.
-
Citation Rate and Query Length — E030 Pre-Registration | The GEO Lab
Read more →: Citation Rate and Query Length — E030 Pre-Registration | The GEO LabE030 pre-registers the hypothesis that Perplexity citation rate increases with query length. A frozen 45-query set across 3 length tiers, 3 stably-cited pages, 5-day measurement window starting 2026-05-19.
-
Claude Leads Perplexity. Ahrefs Shifts to AI Memory Brand. The May 2026 GEO Brand Citation Index Is Out.
Read more →: Claude Leads Perplexity. Ahrefs Shifts to AI Memory Brand. The May 2026 GEO Brand Citation Index Is Out.May 2026 GEO Brand Citation Index: Ahrefs returns to AI Memory Brand with −39.13 delta. Claude hits +60.87 — highest in three runs. SEO vertical has three simultaneous AI Memory…
-
FAQPage Schema and AI Retrieval: 480 Queries, Null Result
Read more →: FAQPage Schema and AI Retrieval: 480 Queries, Null ResultTested FAQPage schema impact on AI retrieval across 480 queries. Result: schema alone produced no measurable citation improvement.
-
Gemini Referenced Our Site 21 Times Out of 100. It Cited Us Zero Times.
Read more →: Gemini Referenced Our Site 21 Times Out of 100. It Cited Us Zero Times.In 160 Gemini queries, thegeolab.net was mentioned in 21.2% of responses and cited in 0%. Three platforms, three completely different visibility mechanisms. Any dashboard that aggregates them into a single…
-
E025 Entity Reinforcement Intervention: Null With Confound
Read more →: E025 Entity Reinforcement Intervention: Null With ConfoundPre-registered entity-graph intervention — ORCID, Zenodo DOIs, public repos, Wikidata — produced no T2 Perplexity citation lift across two monthly measurements. The authority gate held. A 32pp T1 drop at…
-
E016: The Noise Floor Experiment — What AI Citation Rate Looks Like With No Intervention
Read more →: E016: The Noise Floor Experiment — What AI Citation Rate Looks Like With No InterventionE016 ran the 30-check protocol for five consecutive days with zero content changes. The noise floor is 13.3%–20.0% combined. Perplexity: 40%, zero variance. ChatGPT: 0%–20%. Google AIO: absent.
-
We Ran a Controlled Entity Density Experiment. Here’s the Null Result and What It Means.
Read more →: We Ran a Controlled Entity Density Experiment. Here’s the Null Result and What It Means.A controlled experiment measuring entity density vs AI citation rate across 10 pages. Result: 0% competition citation rate. Domain authority gates citation eligibility — entity density operates within that gate.
-
What 1M+ AI Citations Tell Us About Content Strategy in 2026
Read more →: What 1M+ AI Citations Tell Us About Content Strategy in 2026Peec AI and Wix analysed 1M+ AI citations across ChatGPT, Google AI Mode, and Perplexity. Intent beats industry. Listicles own commercial queries. Perplexity values discussions at 2x the rate of…