Context
A 2026 preprint proposes a two-stage GEO measurement framework separating citation selection from citation absorption and evaluates 602 controlled prompts across ChatGPT, Google AI Overview/Gemini and Perplexity using a public dataset.
What changed
The study operationalizes two different outcomes: whether an AI search system searches for and selects a source, and whether a cited page materially contributes language, evidence, structure or factual support to the generated answer.
Why it matters
Citation counts alone can obscure how much a selected source actually influences an answer. Separating selection from absorption provides a more precise measurement model for AI visibility experiments.
Visibility OS interpretation
Visibility OS interprets this as evidence for measuring source selection and answer-level influence separately. A citation should not automatically be treated as equivalent to meaningful absorption into the generated answer.
What remains uncertain
How stable are citation-absorption differences as production models and retrieval stacks change? Which page features causally affect absorption rather than merely correlate with it? How should absorption be normalized across answer lengths and citation styles?