What's Inside the Audit
Sixteen Modules. One Topical-Authority State.
Every module checks a different layer of how Google reads your site as an entity. Together they describe and grade your topical authority across the framework Koray Tugberk built from 85 lectures of semantic SEO methodology.
How your site reads as one entity across all pages - central entity, attribute coverage, frame semantics, and the lexicon Google uses to classify your category.
Topical Map
We map your site against the topical map your category demands - central entity, source context, and central search intent. Then we score the depth of your core section (where you monetize) and the breadth of your outer section (where you prove relevance), node by node.
- Central entity + source context unification check
- Core section depth (primary attribute family)
- Outer section breadth (non-commercial attributes)
- Quality-node candidates for homepage promotion
Missing Entities
Every topical map has an attribute set Google expects you to cover. We compare yours to the category's authority sites and surface the entities competitors talk about that your content doesn't - ranked by ranking-impact potential.
- Attribute filtration: prominence × popularity × relevance
- Competitor entity-coverage benchmark
- Gap list ranked by ranking-impact potential
- Entity-to-page assignment map
Frame Semantics
Queries don't live in dictionaries - they live in user behavior. We extract the predicates, attributes, and functional words your audience actually uses, then check whether your content uses the same semantic frame Google associates with your topic.
- Predicate analysis (how users describe actions)
- Topical borders from user-behavior clustering
- Functional-word coverage (for, to, in, around)
- Three-column query analysis (outrank / classify-with / phrase taxonomy)
Lexical Coverage
Site-wide n-grams and term frequency distribution determine how Google reads your topical authority across pages. We measure the consistency of your phrases and the distribution of your terms against the topic's expected lexicon.
- Site-wide n-gram consistency check
- Term frequency distribution per page + site-wide
- Lemmatization variation per page
- Measurement-unit + entity-name diversity
Page-level and site-level content engineering - heading flow, opening sentences, content briefs, and the 200+ algorithmic authorship rules that decide whether each page gets classified or skipped.
Page Audit
Per-URL diagnostic: heading hierarchy, contextual vector, declaration → evidence flow, bold placements, list formatting, subordinate text, and the make-or-break first sentence after each heading.
- Contextual hierarchy (H2 → H3 → H4 flow)
- Subordinate text quality (first sentence after each heading)
- Declaration → evidence pattern detection
- Bold placement + list-format compliance
Content Plan
A prioritized rewrite + new-content roadmap built from your topical map. Every brief carries a contextual vector, a question set, a contextual structure (table / list / paragraph), and internal-link assignments - ready for a writer to execute.
- Question generation (4 types: definitional, comparative, grouping, boolean)
- Question ranking + grammar normalization
- Contextual vector (straight-line heading flow)
- Per-section format spec (entities, attributes, units)
Content Config
Content configuration is always-on. Every core update gives you new ranking feedback; the configuration adjusts. We identify which pages need to be reconfigured at the next core update - not patched, restructured.
- Ranking feedback per URL across the last 3 core updates
- Semantic-distance shift detection
- Reconfiguration priority list (rewrite / prune / merge)
- Quality-threshold gap per URL
Extractive Answers
AI Overviews and Google's predictive retrieval evaluate the FIRST sentence after each heading. We check whether each section opens with an extractive, declarative answer or buries the lede behind preamble - the difference between a citation and an ignore.
- Answer-first opening sentence audit
- Abstractive vs extractive summary discordance
- Featured-snippet + AI-citation readiness
- Modal-verb certainty audit (be certain, not maybe)
Rule Violations
Koray's 200+ algorithmic authorship rules - be certain, give examples, use numeric values, condition before declaration, POS consistency in lists. We run every page against every rule and flag the violations that hurt classification most.
- Modal-verb certainty + numeric-value presence
- Plural-noun → example pattern
- Part-of-speech consistency in lists
- Condition-before-declaration ordering
Internal-link flow, anchor-text distribution, three-layer architecture, and the cannibalization clusters that confuse Google about which page should rank.
Internal Linking Plan
Your contextual connections drive PageRank distribution. We map every internal link by anchor text, target page, and source position - then build the link plan that makes Google understand which pages are quality nodes.
- Anchor text matching the target page's title
- Maximum 3 uses per anchor across the site
- Link-flow templates per entity type
- Source-position weighting (first link in body matters most)
Link Architecture
Three-layer root → seed → node hierarchy. Quality nodes link from the homepage; outer-section pages link to core-section pages. We map your current architecture against the topical map's intended flow and surface the mismatches.
- Root / Seed / Node tier classification
- Quality-node identification + homepage linkage
- Crawl-depth measurement per page
- Outer-section → core-section link audit
Anchor Text Audit
Anchor text duplication = topical confusion. We measure your anchor distribution per page and across the site, flagging over-optimization, exact-match clusters, and the distance between repeat anchors.
- Anchor frequency + duplication report
- Branded / partial-match / URL distribution
- Distance between repeat anchors
- Anchor-to-target-title alignment
Cannibalization
Cannibalization happens when two pages compete for the same query and Google can't decide which to rank. We surface every cluster of overlapping intent and recommend merge / consolidate / differentiate per cluster.
- Multi-URL query overlap clusters
- Intent overlap (commercial vs informational)
- Merge / consolidate / differentiate recommendation
- Lost-rank attribution to cannibalization
Composite scoring across all 121 concepts, GSC data overlay on the topical map, and AI-Overview citation readiness - what your audit total looks like, why, and what changes when you fix it.
Audit Scores
121 concepts across 6 modules - 57 quantitative + 64 qualitative - score every URL on the framework. You see exactly where each page sits, what's failing, and which fixes will move it most.
- Per-URL score across 6 framework modules
- Quantitative vs qualitative breakdown
- Severity-tier classification per finding
- Priority-fix list ranked by ranking-impact potential
GSC Data
Google Search Console tells the truth about your topical state - new queries, lost queries, shared queries with competitors. We pull your GSC data and overlay it on the topical map to show you where authority is consolidating and where it's leaking.
- Lost-query detection across the last 90 days
- New-query acquisition rate
- Shared-query overlap with the top 3 competitors
- Ranking-shift attribution (core update vs content change)
AI & Compounding
AI Overviews appear in up to 60% of US queries. Compounding authority compounds across both traditional rankings AND AI citations. We check your readiness for both and ensure your content gets extracted the way Google's predictive retrieval expects.
- AI Overview citation readiness per URL
- Structured data coverage (Article / FAQ / HowTo / Product)
- Entity-to-knowledge-graph alignment
- Compounding-authority signal audit (vastness, depth, momentum)
How It Works
From Crawl to Reconfiguration in Four Steps
Define the Anchors
We define your central entity, source context, and central search intent - the three anchors every downstream module references. No audit module runs until these are locked.
Run All 16 Modules
Crawl + GSC pull + competitor scrape feed every module simultaneously. Each URL gets scored against 121 concepts; topical-map alignment is measured node by node.
Synthesize the Findings
Findings are clustered by ranking-impact potential, not severity. You get a single prioritized list - "fix these 12 first, in this order, here's the predicted lift."
Deliver the Reconfig Plan
Audit report + per-URL action items + content brief templates + link plan. Your team can execute, or we can pair the audit with a Monthly SEO engagement to ship the rewrites.
Is This Right for You?
A Semantic Audit Is for Sites That Want to Be Classified, Not Just Crawled
Good Fit
- Publishing content but not gaining topical authority
- Lost rank at a recent core algorithm update
- Competitors outrank you with less content
- Want to be cited by AI Overviews, not just listed
- Building in YMYL or competitive verticals
- Ready to reconfigure pages, not just patch them
Not a Fit
- Want a generic SEO audit checklist PDF
- Have under 20 pages of indexed content
- Expect rankings to move without rewriting any pages
- Think semantic SEO is "just keywords + entities"
- Not ready to act on the reconfiguration plan
Why Semantic SEO Audit
- Built on Koray Tugberk's framework, not generic SEO checklists
- 121 concepts scored per URL - 57 quantitative + 64 qualitative
- Surfaces the entity, attribute, and query gaps blocking AI citations
- Output is an execution-ready rewrite + content plan, tied to revenue
FAQ
Common questions about semantic seo audit.
What is semantic SEO, exactly?
Semantic SEO is the practice of organising your content so a search engine can read your site as a single entity with attributes, predicates, and a knowledge state - not just a collection of pages. The Koray Tugberk framework, which this audit is built on, treats topical authority as the OUTCOME of a well-configured topical map + content network + knowledge base, scored across 121 measurable concepts.
How is this different from a technical SEO audit?
A technical audit checks whether your site is crawlable, indexable, and fast - the plumbing. A semantic SEO audit checks whether the content Google does index gets CLASSIFIED with the topic's authority signal. Most underperforming sites are technically fine but semantically incoherent. Many clients run both audits - technical first, semantic second.
What do I get back?
A scored report across all 16 modules, a per-URL action list ranked by ranking-impact potential, content brief templates for the rewrites, and an internal-link plan. Plus a 60-minute walkthrough call so the priorities and methodology are clear before your team starts executing.
How long does the audit take?
10 to 15 business days from crawl access to delivery. Larger sites (5,000+ URLs) or international setups need additional time. We confirm the exact timeline at intake.
Do I need to know Koray's framework to use the report?
No. Every finding is written in plain English with the why-it-matters, the fix, and the priority. The 60-minute walkthrough call covers the framework concepts that matter for your specific site so your team has the context to execute confidently.
Can you implement the rewrites for us?
Yes. After the audit you can hand it to your team, give it to your devs + writers, or pair it with our Monthly SEO service so we ship the reconfiguration. Most clients choose to have us handle the first wave of rewrites then transition to in-house.
What access do you need?
Read-only Google Search Console access, Google Analytics access, CMS read access (or an export), and a list of your top 5 competitors. Optional but useful: server log access for crawl-budget analysis.
Ready to Move Revenue?
Pick the package that fits your goals or book a free 20-minute growth audit and we will scope the right next step together.