Semantic SEO with Koray’s framework
Source context to internal link in 16 stages, as modelled in the semantic package. Each lesson quotes the enrichment item that defines the stage, in Swedish as written, with its tier.
38 lessons in 6 modules · for content strategists and editors building topical authority · each lesson shows its evidence tier or its basis
Academy Semantic SEO with Koray’s framework
- 01Context4
- 02Entities9
- 03Relations2
- 04Demand6
- 05Content7
- 06Quality thresholds: 10 checks before a page ships10
Open a lesson to read what to do, why, and how the product measures it.
Module 1 of 6
Context
What the site is a source about, in which field, and in which settings.
What to do: Decide who Google should think the site is: one central entity and one perspective that frames every page, product and post.
Why (documented): The 2024 Content Warehouse leak lists site-level focus signals (siteFocusScore, siteRadius). Sites that stay inside one topic are easier to classify.The signal exists per the 2024 API leak, trial testimony or a Google patent. Weight unknown.
How AwesomeSEO measures it: Topic-drift score: semantic distance of each new page from the source context.
Topical mapWhat to do: Source context. What the site is a source about, and from which perspective. Define it once for the site, then let every page reference it.
Why (documented): 13 of the 496 enrichment items serve this stage. The defining item rests on the practice “Define the source context before the first page” (documented).The signal exists per the 2024 API leak, trial testimony or a Google patent. Weight unknown.
8. Source context: Koppla sidans perspektiv till webbplatsens huvudentitet, verksamhet, erbjudande och redaktionella identitet.
How AwesomeSEO measures it: The semantic model holds this stage as a typed object; the validators flag documents that skip it (missing main entity, context dilution, orphan links).
Semantic modelWhat to do: Knowledge domains. The broad field of knowledge the site belongs to. Define it once for the site, then let every page reference it.
Why (practitioner): 16 of the 496 enrichment items serve this stage. A working rule from the semantic SEO framework, tested on your own pages.A working heuristic from experienced SEOs. We test it on your own data.
467. Knowledge domains: Separera bredare kunskapsdomäner från contextual domains; ett knowledge domain kan innehålla flera entiteter och contexts.
How AwesomeSEO measures it: The semantic model holds this stage as a typed object; the validators flag documents that skip it (missing main entity, context dilution, orphan links).
Semantic modelWhat to do: Contextual domains. A setting in which the same entity gets its own attributes and predicates. Define it once for the site, then let every page reference it.
Why (practitioner): 28 of the 496 enrichment items serve this stage. A working rule from the semantic SEO framework, tested on your own pages.A working heuristic from experienced SEOs. We test it on your own data.
466. Contextual domains: Definiera separata sammanhang där samma entitet får olika attribut och predicates.
How AwesomeSEO measures it: The semantic model holds this stage as a typed object; the validators flag documents that skip it (missing main entity, context dilution, orphan links).
Semantic modelModule 2 of 6
Entities
Central entity, main entity per document, attributes and their values.
What to do: Central entity. The entity that unites everything the site publishes. Define it once for the site, then let every page reference it.
Why (practitioner): 18 of the 496 enrichment items serve this stage. The defining item rests on the practice “Map root, seed and outer nodes before publishing” (practitioner).A working heuristic from experienced SEOs. We test it on your own data.
273. Central entity: Definiera den centrala entitet som förenar webbplatsens innehåll.
How AwesomeSEO measures it: The semantic model holds this stage as a typed object; the validators flag documents that skip it (missing main entity, context dilution, orphan links).
Semantic modelWhat to do: Central versus main entity. What one document is about. Exactly one per document. Define it once for the site, then let every page reference it.
Why (practitioner): 70 of the 496 enrichment items serve this stage. A working rule from the semantic SEO framework, tested on your own pages.A working heuristic from experienced SEOs. We test it on your own data.
494. Central vs main entity: Gör explicit skillnad mellan webbplatsens central entity och varje dokuments main entity.
How AwesomeSEO measures it: The semantic model holds this stage as a typed object; the validators flag documents that skip it (missing main entity, context dilution, orphan links).
Semantic modelWhat to do: Core attributes. A property of an entity people ask about or decide on. Define it once for the site, then let every page reference it.
Why (practitioner): 69 of the 496 enrichment items serve this stage. A working rule from the semantic SEO framework, tested on your own pages.A working heuristic from experienced SEOs. We test it on your own data.
77. Grundläggande attribut: Beskriv de egenskaper som krävs för att förstå och bedöma huvudentiteten.
How AwesomeSEO measures it: The semantic model holds this stage as a typed object; the validators flag documents that skip it (missing main entity, context dilution, orphan links).
Semantic modelWhat to do: Entity–attribute–value facts. The fact for an attribute, with unit, source and date. Define it once for the site, then let every page reference it.
Why (practitioner): 119 of the 496 enrichment items serve this stage. The defining item rests on the practice “Model Entity → Attribute → Value and place attributes by P-R-P” (practitioner).A working heuristic from experienced SEOs. We test it on your own data.
82. EAV: Strukturera fakta som entitet → attribut → värde, exempelvis tjänst → månadsavgift → belopp.
How AwesomeSEO measures it: The semantic model holds this stage as a typed object; the validators flag documents that skip it (missing main entity, context dilution, orphan links).
Semantic modelWhat to do: Name the real things (products, places, people, models, standards) and describe them through their attributes.
Why (measured): Number of Distinct Entities Used is the most durable strong factor: in the top 200 for 81 of 93 months, average r −0.27.Correlates with position in the TopSEOFactors data. Correlation, not causation.
How AwesomeSEO measures it: Entity count and entity-in-title/H1/H2 against page-1 averages.
- Number of Distinct Entities Used #45 · r −0.18
- Entities in Title Tag #73 · r −0.17
- Entities in H2 Tags #44 · r −0.18
What to do: Score each attribute 1–5 for prominence, relevance and popularity. 13+ gets its own H2 and a filter, 10–12 an H3, 7–9 a paragraph, below 7 a fact-card line.
Why (practitioner): Turns "cover the topic" into a repeatable page template, and the same model drives facets, schema and fact cards.A working heuristic from experienced SEOs. We test it on your own data.
How AwesomeSEO measures it: EAV editor with automatic placement and schema property mapping.
Topical mapWhat to do: Reference each page’s main entity by a stable @id in JSON-LD and link it to Wikidata, Wikipedia and official profiles with sameAs.
Why (confirmed): Google’s structured-data guidelines support sameAs and consistent identifiers for organisations and people.Google states it in documentation or official guidance.
How AwesomeSEO measures it: Entity graph with @id coverage and sameAs completeness.
Topical mapWhat to do: Cover roughly 60–70% of the attributes the top results agree on, then add attributes and facts nobody else has.
Why (documented): Google holds a granted patent on information-gain scoring (2022). Agreeing with the consensus shows alignment; new attributes justify ranking you.The signal exists per the 2024 API leak, trial testimony or a Google patent. Weight unknown.
How AwesomeSEO measures it: Attribute coverage vs top 20: consensus coverage and information-gain share.
SERP LabWhat to do: Compare attributes in other languages’ SERPs and knowledge panels; those missing from Swedish results are first-mover gain.
Why (practitioner): Google understands entities across languages, but local SERPs often lack facts that are common elsewhere.A working heuristic from experienced SEOs. We test it on your own data.
SERP LabModule 3 of 6
Relations
Predicates, frames and the roles around them.
What to do: Frame semantics. The action or relation, and the situation (frame) it implies. Define it once for the site, then let every page reference it.
Why (practitioner): 39 of the 496 enrichment items serve this stage. A working rule from the semantic SEO framework, tested on your own pages.A working heuristic from experienced SEOs. We test it on your own data.
459. Frame semantics: Modellera kompletta situationer kring handlingar. `Köpa` implicerar exempelvis köpare, säljare, objekt, pris, betalning och överföring.
How AwesomeSEO measures it: The semantic model holds this stage as a typed object; the validators flag documents that skip it (missing main entity, context dilution, orphan links).
Semantic modelWhat to do: FrameNet roles. Who acts, on what, with what, where, when and with which result. Define it once for the site, then let every page reference it.
Why (practitioner): 41 of the 496 enrichment items serve this stage. A working rule from the semantic SEO framework, tested on your own pages.A working heuristic from experienced SEOs. We test it on your own data.
460. FrameNet-roller: Definiera vilka semantiska roller som normalt behöver fyllas inom varje frame och koppla dem till Semantic Role Labeling, SRL.
How AwesomeSEO measures it: The semantic model holds this stage as a typed object; the validators flag documents that skip it (missing main entity, context dilution, orphan links).
Semantic modelModule 4 of 6
Demand
Themes, templates, queries and the questions behind them.
What to do: Query themes. The underlying need a group of queries shares. Define it once for the site, then let every page reference it.
Why (practitioner): 10 of the 496 enrichment items serve this stage. A working rule from the semantic SEO framework, tested on your own pages.A working heuristic from experienced SEOs. We test it on your own data.
439. Query themes: Gruppera queries efter det bakomliggande semantiska temat, inte enbart efter keyword cluster.
How AwesomeSEO measures it: The semantic model holds this stage as a typed object; the validators flag documents that skip it (missing main entity, context dilution, orphan links).
Semantic modelWhat to do: Query templates. A reusable search pattern with slots. Define it once for the site, then let every page reference it.
Why (practitioner): 24 of the 496 enrichment items serve this stage. A working rule from the semantic SEO framework, tested on your own pages.A working heuristic from experienced SEOs. We test it on your own data.
437. Query templates: Modellera återkommande frågemönster som `hur + VERB + ENTITY`, `ENTITY + pris` och `ENTITY + för + AUDIENCE`, snarare än bara individuella sökord. Koppla predicates till återanvändbara query templates.
How AwesomeSEO measures it: The semantic model holds this stage as a typed object; the validators flag documents that skip it (missing main entity, context dilution, orphan links).
Semantic modelWhat to do: Primary keyword. What people actually type. Define it once for the site, then let every page reference it.
Why (practitioner): 61 of the 496 enrichment items serve this stage. A working rule from the semantic SEO framework, tested on your own pages.A working heuristic from experienced SEOs. We test it on your own data.
12. Huvudsökord: Identifiera den huvudsakliga sökfras som motsvarar sidans ämne och uppgift.
How AwesomeSEO measures it: The semantic model holds this stage as a typed object; the validators flag documents that skip it (missing main entity, context dilution, orphan links).
Semantic modelWhat to do: Query–question–answer link. The question behind the query. Define it once for the site, then let every page reference it.
Why (practitioner): 27 of the 496 enrichment items serve this stage. A working rule from the semantic SEO framework, tested on your own pages.A working heuristic from experienced SEOs. We test it on your own data.
292. Query–question–answer: Koppla sökfrasen till den underliggande frågan och ett uttryckligt svar.
How AwesomeSEO measures it: The semantic model holds this stage as a typed object; the validators flag documents that skip it (missing main entity, context dilution, orphan links).
Semantic modelWhat to do: List the entity class, its shared attributes and the query templates. Attributes with their own query template get a page; the rest become sections.
Why (practitioner): "Query deserves a page" and complete entity-class coverage are the core of Koray’s topical-authority method.A working heuristic from experienced SEOs. We test it on your own data.
Topical mapWhat to do: When several of your URLs swap positions for the same query, pick one primary page, merge or differentiate the others and point internal links at the primary.
Why (practitioner): Competing pages split links and signals and confuse which page should rank; consolidation usually stabilises the winner.A working heuristic from experienced SEOs. We test it on your own data.
How AwesomeSEO measures it: Cannibalisation report from Search Console: queries with two or more ranking URLs and position swaps.
Search performanceModule 5 of 6
Content
Answers, documents and the link to the next query.
What to do: Query–answer format alignment. The explicit answer, in the format the question needs. Define it once for the site, then let every page reference it.
Why (practitioner): 111 of the 496 enrichment items serve this stage. A working rule from the semantic SEO framework, tested on your own pages.A working heuristic from experienced SEOs. We test it on your own data.
493. Query–answer format alignment: Matcha frågetyp mot lämplig svarstyp: `definition → definition`, `jämförelse → jämförbar struktur`, `hur → process`, `hur mycket → kvantifierat svar`.
How AwesomeSEO measures it: The semantic model holds this stage as a typed object; the validators flag documents that skip it (missing main entity, context dilution, orphan links).
Semantic modelWhat to do: Page type. The page that carries the answers. Define it once for the site, then let every page reference it.
Why (practitioner): 132 of the 496 enrichment items serve this stage. The defining item rests on the practice “Serve the page type the SERP shows” (practitioner).A working heuristic from experienced SEOs. We test it on your own data.
6. Sidtyp: Välj exempelvis guide, tjänstesida, produkt, kategori, ortssida, jämförelse, recension, ordlista, nyhet eller verktyg.
How AwesomeSEO measures it: The semantic model holds this stage as a typed object; the validators flag documents that skip it (missing main entity, context dilution, orphan links).
Semantic modelWhat to do: Internal links. The link that takes the reader to the next query. Define it once for the site, then let every page reference it.
Why (confirmed): 26 of the 496 enrichment items serve this stage. Basis: Google Search Central: Link best practices for Google.Google states it in documentation or official guidance.
183. Internlänkar: Koppla sidan till överordnade ämnen, syskonsidor, underkategorier och fördjupningar.
How AwesomeSEO measures it: The semantic model holds this stage as a typed object; the validators flag documents that skip it (missing main entity, context dilution, orphan links).
Semantic modelWhat to do: Open each section with a direct answer of about 40 words, exact where consensus allows, with the answer in bold. Use real questions as H3s.
Why (measured): Questions in headings correlate (QS004, r −0.16, 82 of 93 months) and direct answers are what featured snippets and AI answers lift.Correlates with position in the TopSEOFactors data. Correlation, not causation.
How AwesomeSEO measures it: Answer-first check per heading in the brief and the audit.
- Number of Questions In Headings #76 · r −0.16
What to do: Ship a whole cluster (typically 5–20 pages) at once with its internal links mapped, then submit the sitemap with accurate lastmod.
Why (practitioner): A complete, interlinked cluster shows coverage in one crawl. Google retired the sitemap ping endpoint in 2023, so use Search Console and lastmod.A working heuristic from experienced SEOs. We test it on your own data.
How AwesomeSEO measures it: Cluster completeness: share of planned URLs live, indexed and linked.
Topical mapWhat to do: Pick the next cluster by semantic adjacency to what already ranks (AI tools → AI in business, not AI robots). Keep maintaining what exists.
Why (practitioner): Topical authority transfers between close topics; big leaps start from zero.A working heuristic from experienced SEOs. We test it on your own data.
How AwesomeSEO measures it: Next-cluster ranking: adjacency × demand, leaps flagged.
Topical mapWhat to do: List every post, product, publication and event that mentions the entity, and link people, companies and products to each other through their attributes.
Why (practitioner): Dense, meaningful connections between entity pages form the semantic content network Koray describes.A working heuristic from experienced SEOs. We test it on your own data.
Topical mapModule 6 of 6
Quality thresholds: 10 checks before a page ships
From Koray Tuğberk Gübür, semantic SEO course material, as a checklist per page type.
What to do: At least one novel, exclusive or rare attribute that the top 20 miss.
Why (practitioner): A working rule from Koray Tuğberk Gübür, semantic SEO course material. Shown as practitioner tier: coherent and testable, not a confirmed ranking signal.A working heuristic from experienced SEOs. We test it on your own data.
How AwesomeSEO measures it: The content module runs the checklist for the page type before a draft is sent for approval.
ContentWhat to do: Numbers with units, not “many”, “often” or “fast”.
Why (practitioner): A working rule from Koray Tuğberk Gübür, semantic SEO course material. Shown as practitioner tier: coherent and testable, not a confirmed ranking signal.A working heuristic from experienced SEOs. We test it on your own data.
How AwesomeSEO measures it: The content module runs the checklist for the page type before a draft is sent for approval.
ContentWhat to do: The answer to the main question in one passage of 40 words or fewer.
Why (practitioner): A working rule from Koray Tuğberk Gübür, semantic SEO course material. Shown as practitioner tier: coherent and testable, not a confirmed ranking signal.A working heuristic from experienced SEOs. We test it on your own data.
How AwesomeSEO measures it: The content module runs the checklist for the page type before a draft is sent for approval.
ContentWhat to do: Each question answered in one definitive sentence before any detail.
Why (practitioner): A working rule from Koray Tuğberk Gübür, semantic SEO course material. Shown as practitioner tier: coherent and testable, not a confirmed ranking signal.A working heuristic from experienced SEOs. We test it on your own data.
How AwesomeSEO measures it: The content module runs the checklist for the page type before a draft is sent for approval.
ContentWhat to do: Entity, attribute and value in every section: “the sheet weighs 165 g/m²”.
Why (practitioner): A working rule from Koray Tuğberk Gübür, semantic SEO course material. Shown as practitioner tier: coherent and testable, not a confirmed ranking signal.A working heuristic from experienced SEOs. We test it on your own data.
How AwesomeSEO measures it: The content module runs the checklist for the page type before a draft is sent for approval.
ContentWhat to do: Every plural (“sizes”, “colours”) is followed by concrete examples.
Why (practitioner): A working rule from Koray Tuğberk Gübür, semantic SEO course material. Shown as practitioner tier: coherent and testable, not a confirmed ranking signal.A working heuristic from experienced SEOs. We test it on your own data.
How AwesomeSEO measures it: The content module runs the checklist for the page type before a draft is sent for approval.
ContentWhat to do: List items start with the same word class.
Why (practitioner): A working rule from Koray Tuğberk Gübür, semantic SEO course material. Shown as practitioner tier: coherent and testable, not a confirmed ranking signal.A working heuristic from experienced SEOs. We test it on your own data.
How AwesomeSEO measures it: The content module runs the checklist for the page type before a draft is sent for approval.
ContentWhat to do: A number, a source or your own test follows every claim.
Why (practitioner): A working rule from Koray Tuğberk Gübür, semantic SEO course material. Shown as practitioner tier: coherent and testable, not a confirmed ranking signal.A working heuristic from experienced SEOs. We test it on your own data.
How AwesomeSEO measures it: The content module runs the checklist for the page type before a draft is sent for approval.
ContentWhat to do: A declared cadence for meaningful updates: monthly, quarterly or annual.
Why (practitioner): A working rule from Koray Tuğberk Gübür, semantic SEO course material. Shown as practitioner tier: coherent and testable, not a confirmed ranking signal.A working heuristic from experienced SEOs. We test it on your own data.
How AwesomeSEO measures it: The content module runs the checklist for the page type before a draft is sent for approval.
ContentWhat to do: One main entity and one intent per page.
Why (practitioner): A working rule from Koray Tuğberk Gübür, semantic SEO course material. Shown as practitioner tier: coherent and testable, not a confirmed ranking signal.A working heuristic from experienced SEOs. We test it on your own data.
How AwesomeSEO measures it: The content module runs the checklist for the page type before a draft is sent for approval.
ContentNext track
Practise on the demo workspace: Semantic model.
Example data for a fictional Swedish brand, every number with its source, interval and n.