Skip to content

Page deep dive

Full diagnostic for one URL using Google indexation, search performance, Core Web Vitals and on-page evidence. When Bing is configured, add its page-query and URL traffic data without turning Bing URL fields into an indexation verdict.

Replace yourdomain.com/your-page-path with the actual URL you want to diagnose.


Give me a full diagnostic of yourdomain.com/your-page-path. Start with the indexing status, then performance, then Core Web Vitals.

The assistant calls inspect_url, page_health_score, get_search_by_page_query and crux_page_vitals in sequence and returns the available evidence with the health-score breakdown.


Is Google actually indexing this page? When was it last crawled and what was the result?

Which queries bring visitors to this page? Are there any with high impressions but low CTR that I should rewrite the title or meta description for?

How are the Core Web Vitals on this page trending over the last 25 weeks? Is LCP or INP above the threshold?

Based on the queries this page already ranks for and the ones where it gets impressions but no clicks, what topics should I cover or expand?

The assistant calls content_brief to combine the page’s top GSC queries, detected question queries and optional GA4 engagement data. Inspect the page separately before turning those observations into content recommendations.


I think I should rank higher for [query] on this page. What position and CTR are actually observed, and how do they compare with this page’s other measured queries?

The assistant uses get_search_by_page_query for the observed query-level values. The exposed tools do not provide a query-level expected-CTR benchmark, so any external benchmark must be named and treated as a heuristic.

Bing is configured for https://yourdomain.com/. Add bing_page_query_stats, bing_url_traffic and bing_url_info for this exact page. Keep Bing position and freshness fields separate from Google, and mark unavailable fields as unknown.

Use compare_search_engines only when you want a bounded side-by-side comparison. It omits deltas when exact equal observed windows are not established.