Keywords are how we search. Entities are how Google understands. This portal explores the shift from keyword matching to conceptual understanding — and what it means for every business trying to be found online.
Imagine you type "best Italian near me" into Google. You don't get pages that contain those exact words. You get restaurants. Google knows what you mean, not just what you typed.
That gap between typed words and understood meaning is where entity-based search lives. Google has been building a map of the real world — businesses, people, places, concepts — and connecting them to each other. This map is called the Knowledge Graph.
Your business either exists on that map or it doesn't. This portal explains how the map works, what it takes to appear on it, and why the rules of search visibility have quietly changed beneath everyone's feet.
Google's Knowledge Graph contains billions of facts about entities and their relationships.
Each area of this portal is grounded in published research, patent filings, and documented Google behavior. No speculation, no optimization sales pitches.
Google's Knowledge Graph is a structured database of entities and their relationships. It's not a search index — it's a semantic network that lets Google reason about what things are, not just where words appear.
Read the guideA Wikipedia mention doesn't just add a backlink. It signals to Google that an entity is notable enough to be documented by an independent, editorial source. That signal carries unusual weight in how Google categorizes businesses.
Understand the signalSchema markup is structured data embedded in a webpage that explicitly tells Google what type of thing a page is about. It's the bridge between human-readable content and machine-readable entity data.
See the configurationsGoogle has filed patents describing how it identifies, disambiguates, and scores entities. Reading those patents reveals the actual mechanisms behind search behavior — not guesswork, not SEO myth.
Explore the patentsLocal businesses face a specific challenge: Google must understand not just that they exist, but what category they belong to, where they serve, and how they relate to the surrounding area. Entity clarity directly affects local pack visibility.
Start from the basicsSearch is not static. Google's published research papers, patent applications, and public statements point toward where entity understanding is heading. This portal tracks those signals and translates them into readable editorial content.
Browse by complexityThis is how Google moves from raw web content to a structured understanding of what your business actually is.
Googlebot crawls web pages and collects raw text, structured data, and link relationships. Every mention of your business name, address, or category is recorded.
Natural language processing identifies named entities in text — organizations, locations, people, products. Each entity gets a candidate ID within Google's systems.
Multiple signals determine which entity a mention refers to. Patents describe salience scoring — how central an entity is to a document — which affects how confidently Google associates content with your business.
Confirmed entities are connected to existing Knowledge Graph nodes or new nodes are created. Relationships between entities — "located in," "founded by," "serves" — are mapped and weighted.
Entity understanding shapes what appears in Knowledge Panels, local packs, featured snippets, and voice search answers. The clearer your entity, the more precisely Google can match you to relevant queries.
Most SEO content is written by people interpreting other people's interpretations. This portal goes to primary sources: Google's actual patent filings, published research papers from Google Brain and DeepMind, and documented API behaviors.
When Google filed patents describing "entity salience" — a measure of how central an entity is to a document — it revealed something important. Google doesn't just detect that your business name appears on a page. It measures how much that page is fundamentally about your business versus merely mentioning it in passing.
That distinction changes how you think about content entirely. A page that mentions your business name twenty times in passing is less valuable than a page where your business is the undisputed subject.
Most people arrive here with a vague sense that "SEO has changed." Here's the path from that feeling to actual understanding.
Something changed in your search rankings. Or you read something about entities and Knowledge Graphs. You're not sure what it means for your situation. That's the right place to start.
Browse by complexity to find content matched to your current knowledge. Beginner articles explain the fundamentals. Intermediate pieces dig into mechanisms. Advanced content engages with the actual research.
Every claim on this portal links to its source. Follow those links. Read the patents. Scan the research papers. The goal isn't for you to trust us — it's for you to verify and understand independently.
Understanding entity-based search is less about tactics and more about developing an accurate mental model of how Google reasons. Once that model clicks, a lot of previously confusing SEO behavior makes sense.
Entity-based search is not a finished system. Google publishes new research regularly. This portal tracks those developments and adds editorial content as the landscape changes. Bookmark it and come back.
A multinational corporation has thousands of web pages, press mentions, Wikipedia articles, and Wikidata entries reinforcing its identity as an entity. Google has abundant signal.
A local plumber in Atlanta has a Google Business Profile, a website, and maybe a few directory listings. The entity signal is thin. That thinness is exactly why understanding how to reinforce entity clarity matters disproportionately for local businesses.
When Google understands your business as a clear, well-defined entity — its category, location, services, and relationships to the surrounding area — local search visibility follows more naturally. Not through tricks. Through clarity.
Read Local-Focused Articles
Patents
The patent describes a scoring system that measures how central an entity is to a document. Here's a plain-English walkthrough of the key claims and what they imply for content structure.
Read analysis
Wikipedia
There's a difference between having a Wikipedia article and being mentioned in one. Both send signals to Google, but they operate through different mechanisms and carry different weight in entity recognition.
Read analysis
Schema
Not all schema types are equal in their entity-feeding potential. This guide maps the schema.org vocabulary to the entity types Google's Knowledge Graph actually tracks, with attention to local business types.
Read guideThis portal is maintained by researchers in Atlanta, GA. If you have a question about a specific topic, a research source you'd like us to examine, or feedback on our coverage, reach out directly.
Get in Touch