CONCEPT

Entity-Based SEO

Properties1
tagsentity-based-seo, entities, knowledge-graph, transformers, local-seo, ai-search

3 min read

Entity-Based SEO

Optimizing for how Google understands entities — real businesses, services, locations, and the relationships between them — rather than matching keywords.

Definition

Entity-based SEO reflects Google’s shift (accelerated by the transformer architecture Google introduced in 2017) from keyword matching to understanding entities and their relationships. In Caleb Ulku’s framing, a local business’s Google Business Profile, its website, each listed service and category, and its geography are all entities; Google constantly checks whether they align, and the closer they match, the more it trusts and ranks the business. This is why a business with 12 reviews can outrank one with 5,000 — Google is evaluating entity connections (business ↔ service ↔ location), not just counting keywords or reviews.

The practical upshot: you rank by explicitly declaring your entity relationships (via the GBP, GBP–Website Alignment (Consistency Signals), and the Core 30 page structure) rather than hoping Google infers them from blog content. The same entity signals are increasingly read by AI models (see GEO / AEO (Getting Recommended by AI)), so entity alignment is a single lever for both classic and AI search.

Key Properties

  • Entities > keywords: Google matches business/service/location entities and their relationships (Caleb dates the shift to ~2018 in ranking behavior; the transformer itself is 2017).
  • The GBP is Google’s canonical representation of the business entity — optimizing it directly defines the entity.
  • Alignment/consistency across GBP, website, and content is the ranking lever (GBP–Website Alignment (Consistency Signals), Core 30).
  • Explains review-count upsets (12 reviews outranking 5,000) — connections, not counts.
  • Convergence: ChatGPT, Perplexity, Claude evaluate similar entity signals → fixing it for Google helps AI recommendation.

Examples from Sources

ExampleSource
Tool counts how many local GBPs use each category to choose aligned secondary categoriesThis AI Tool Will Make You RANK FAST (Full Walkthrough)
Neighborhood/landmark pages that connect the business entity to a specific location entityThis will rank you #1 AND bring you customers (a.k.a. 'I Spent 10+ Years Ranking Businesses #1. This Ranks the FASTEST')

In the Sources

SourceContext
This AI Tool Will Make You RANK FAST (Full Walkthrough)”Since 2018, Google doesn’t match keywords, it matches entities” — the clearest statement.
This will rank you #1 AND bring you customers (a.k.a. 'I Spent 10+ Years Ranking Businesses #1. This Ranks the FASTEST')Introduces the ~2018 transformer shift to explain why hyper-local pages rank fast.
The BEST SEO Tutorial for Businesses 2026 (Full Guide) (saved as 'The BEST Local SEO Tutorial for Beginners 2026')”Google invented the transformer in 2017; the T in ChatGPT stands for transformer.”
2026 SEO Secrets That Ranked 1,000+ Sites (James Dooley, interviewed by Sterling Sky)Independent corroboration + a concrete authoring technique: Semantic Triples (Entity–Attribute–Value) (entity–attribute–value).

Concepts: Semantic Triples (Entity–Attribute–Value), GBP–Website Alignment (Consistency Signals), Core 30, Local Ranking Factors (Relevance, Proximity, Prominence/Authority), GEO / AEO (Getting Recommended by AI), Schema Markup (Structured Data) Entities: Google, Google Business Profile, James Dooley