Content Optimizer flags relevance gaps and feeds those low-performing queries into a Bayesian testing loop that benchmarks the target page against top-ranking SERP competitors. After a generative ideator drafts targeted line-by-line patch edits and a judge model scores the revisions across multiple rounds, the winning modifications and underlying rank factors are synthesized into actionable, line-referenced editorial briefs.