What Makes a Page Citable: Inside the GEO-16 Page-Quality Study

The study in one paragraph

GEO-16 (arXiv:2509.10762, 2025) is a 16-pillar auditing framework that converts on-page quality signals into banded pillar scores and a normalised GEO score from 0 to 1. The authors used 70 product-intent prompts, collected 1,702 citations across Brave Summary, Google AI Overviews and Perplexity, and audited 1,100 unique URLs, then tested which pillars predicted actually being cited. Read it: https://arxiv.org/abs/2509.10762

How it worked

Sixteen pillars were scored per page and banded. The team then modelled which pillars were associated with citation, using logistic models with domain-clustered standard errors.

What predicted citation

Pillars related to metadata and freshness, semantic HTML, and structured data showed the strongest association with being cited. The authors also report practical operating points: a normalised score of at least 0.70 combined with at least 12 pillar hits aligned with substantially higher citation rates in their corpus.

In plain terms, how machine-readable and how current a page is mattered more than prose alone.

What it does not claim

This is about page quality, not writing style, and it is not evidence that any single editorial tactic guarantees citation. The paper states that the study is observational, focuses on English-language B2B SaaS pages, and it discusses limitations, threats to validity and reproducibility.

Why this matters

Most AI SEO advice is about phrasing. This study points at the unglamorous basics that you can audit and fix: clean semantic HTML, structured data machines can parse, and visible freshness signals.

What to do with it

• Audit the pages you most want cited against the three strongest pillars.
• Add or repair structured data.
• Make the last-updated date real and visible.
• Re-audit quarterly, because freshness is one of the measured pillars.

Sources

• GEO-16, arXiv:2509.10762: https://arxiv.org/abs/2509.10762