Evidence 40

❝Evidence 93 Google must crawl, render and index a page before it can rank Google Search Central documents crawling, rendering and indexing as distinct prerequisite stages before any ranking can occur. Technical SEO ❝Evidence 88 Crawl budget is finite and wasted on low-value URLs Google's large-site crawl budget guidance explains that crawl capacity and demand are limited, so wasted crawling on low-value URLs delays discovery of important pages. Technical SEO ❝Evidence 92 E-E-A-T and Experience are central to Google's quality assessment Google's Search Quality Rater Guidelines centre page and site quality on Experience, Expertise, Authoritativeness and Trust, with Trust as the most important member. Authority & Trust ❝Evidence 85 Rater judgements calibrate algorithms rather than directly ranking pages Google states that quality rater ratings do not directly affect the ranking of any individual page but are used to evaluate and improve search algorithms. Authority & Trust ❝Evidence 90 Valid structured data enables rich result eligibility Google's structured data documentation states that valid schema.org markup makes pages eligible for rich results and enhanced appearances in search. Technical SEO ❝Evidence 88 Core Web Vitals are a measurable page experience signal Google documents Core Web Vitals (LCP, INP, CLS) as real-world measures of loading, interactivity and visual stability that feed its page experience assessment. Technical SEO ❝Evidence 89 Google rewards people-first, satisfying content over search-engine-first content Google's helpful content guidance instructs creators to produce people-first content and warns that content made primarily to rank can underperform site-wide. Content Optimisation ❝Evidence 82 Links are weighted by likelihood of being clicked (Reasonable Surfer) Google's Reasonable Surfer patent describes assigning different weights to links based on the probability a user would actually click them. Technical SEO ❝Evidence 88 PageRank models authority as link-derived probability The original PageRank algorithm models a page's importance as a function of the quantity and importance of pages linking to it. Authority & Trust ❝Evidence 80 Information gain scores reward documents that add new information Google's information gain patent describes scoring documents by how much new information they add beyond what a user has already seen. Content Optimisation ❝Evidence 87 Answer engines retrieve external documents to ground generated answers Retrieval-Augmented Generation research shows that grounding language-model outputs in retrieved documents improves factual accuracy. GEO & AI Search ❝Evidence 82 Passage-level retrieval rewards self-contained, well-scoped chunks Dense passage retrieval research shows retrieval operates at the passage level, favouring self-contained chunks that answer a query directly. GEO & AI Search ❝Evidence 81 AI Overviews surface links to corroborating web sources Google states that AI Overviews are built to highlight and link to supporting web content, sending users to corroborating sources. AI Visibility ❝Evidence 86 Relevance, distance and prominence drive local ranking Google documents that local search results are ranked mainly by relevance, distance and prominence, with reviews contributing to prominence. Commercial SEO ❝Evidence 84 Users scan pages rather than read them (F-pattern) Nielsen Norman Group eye-tracking research shows users scan web content in F-shaped and layer-cake patterns rather than reading word by word. Content Optimisation ❝Evidence 89 Google selects one canonical among duplicate or similar pages Google's canonicalisation documentation explains it groups duplicate URLs and selects a single canonical to index and rank. Technical SEO ❝Evidence 85 Google may rewrite titles and snippets based on relevance Google documents that it generates title links and snippets algorithmically, sometimes rewriting them to better match the query. Content Optimisation ❝Evidence 76 Some queries favour fresh content Industry research and Google guidance indicate that freshness is a relevance factor for queries where recency matters. Content Optimisation ❝Evidence 78 Comprehensive topical coverage correlates with authority and rankings Industry research indicates that sites covering a topic comprehensively via interlinked clusters tend to earn broader visibility than isolated pages. Content Optimisation ❝Evidence 70 SearchScore observation: indexing failures mask content quality Across SearchScore audits, pages with strong content frequently underperform purely because of unresolved indexing or canonical issues. Technical SEO ❝Evidence 68 SearchScore observation: extractable structure raises AI citation likelihood SearchScore observes that pages with clear claims, direct answers and clean structure are cited by AI answer engines more often than dense prose. GEO & AI Search ❝Evidence 74 SearchScore observation: internal linking is the most underused lever SearchScore audits repeatedly find priority pages under-linked internally while boilerplate absorbs most link equity. Technical SEO ❝Evidence 69 SearchScore observation: consensus-restating content plateaus SearchScore observes that pages restating widely available consensus rarely rank or get cited, whereas pages adding original data or experience break through. Content Optimisation ❝Evidence 80 Consistent entity signals strengthen recognition in the Knowledge Graph Structured data and consistent entity references help search and answer engines recognise and disambiguate a brand as a distinct entity. Authority & Trust ❝Evidence 87 Client-side rendered content risks delayed or missed indexing Google documents that JavaScript content is rendered in a deferred second wave, creating risk that unrendered or blocked content is not indexed. Technical SEO ❝Evidence 85 Links are weighted by prominence, not counted equally Google's 'Reasonable Surfer' patent describes weighting a link by how likely a user is to click it. Technical SEO ❝Evidence 78 Content that adds new information can rank higher Google's information-gain patent describes ranking documents by how much new information they add beyond what a user has already seen. Content Optimisation ❝Evidence 88 AI answers are grounded in retrieved source documents The foundational RAG paper shows language models producing answers conditioned on documents retrieved at query time. GEO & AI Search ❝Evidence 84 Language models attend most to the start of a passage The 'Lost in the Middle' study shows models use information best when it appears at the beginning or end of their context. GEO & AI Search ❝Evidence 86 Experience, Expertise, Authoritativeness and Trust guide quality assessment Google's Search Quality Rater Guidelines define E-E-A-T as a lens for judging page and site quality. Authority & Trust ❝Evidence 85 People-first content is Google's stated quality bar Google's guidance on creating helpful content asks publishers to prioritise people over search engines. Content Optimisation ❝Evidence 84 Valid structured data unlocks rich result eligibility Google's structured data documentation explains that valid, guideline-compliant markup makes pages eligible for enhanced results. Technical SEO ❝Evidence 80 Page experience and Core Web Vitals are a quality signal Google's page experience documentation describes Core Web Vitals as part of how it assesses a good user experience. Technical SEO ❝Evidence 83 Crawl budget must be managed on large sites Google's large-site documentation explains how crawl capacity and demand shape what gets crawled and refreshed. Technical SEO ❝Evidence 85 Retrieval matches meaning, not just keywords The Dense Passage Retrieval paper shows embedding-based retrieval outperforming keyword matching for question answering. GEO & AI Search ❝Evidence 84 Duplicate URLs should be consolidated to one canonical Google's documentation explains how it consolidates duplicate pages and why declaring a canonical helps. Technical SEO ❝Evidence 79 Authority flows through the link graph The original PageRank research modelled a page's importance as a function of the links pointing to it. Authority & Trust ❝Evidence 80 Users scan pages rather than read them linearly Nielsen Norman Group eye-tracking research shows people scan web content in F-shaped and layer-cake patterns. Content Optimisation ❝Evidence 62 Answer-first, well-sourced pages are cited more often by AI engines Across SearchScore audits, pages with clear, self-contained answers and visible sourcing appear more often in AI answers. AI Visibility ❝Evidence 60 Brands absent from AI recommendations tend to lack third-party corroboration SearchScore audits repeatedly find that brands omitted from AI recommendations have weak entity signals and little independent mention. AI Visibility