Core concepts
Every score in the product comes from one canonical model. These are the concepts that model exposes.
Content Score
Content Score is a 0–100 measure of how well a draft matches the traditional Google ranking surface for a target keyword. It weights seven signals: term frequency, entity coverage, heading structure, word count, readability, internal links, and GEO signals. No single signal determines the score — each weight reflects its empirical impact on ranking outcomes.
GEO Score
GEO Score is a 0–100 measure of how ready a draft is for AI retrieval, citation, and answer extraction. It weights five signals: Entity Authority, Factual Density, Answer-First Formatting, Source Credibility, and Freshness. Where Content Score targets document ranking, GEO Score targets passage-level extraction by AI search engines.
Review Gate
The Review Gate is a server-enforced checkpoint that blocks publishing of articles with a Content Score below 70/100. Writers receive an ExplainScore breakdown — the ranked list of signals that failed and the specific changes that would raise the score above the threshold.
Action Rail
The Action Rail prioritizes the changes most likely to move a score and explains which signal each action affects, so editors fix the highest-impact issues first instead of guessing.