Old-school SEO was basically “sprinkle the magic keyword dust and hope the Google gods smile.” That party is over. Google now tries to understand what you mean. That is where entities come in.

What’s an entity, really?

Infographic of everyday things that count as entities, from a person and a company to Kilimanjaro, hiking, and pizza.

You don’t need a PhD for this. If you can name it, it is probably an entity: a person, a company, a product, Kilimanjaro, hiking, pizza. Concepts like “fitness” count too. Google groups ideas into entities and connects them. If you put “walking,” “hills,” and “fitness” together, you point to “hiking,” even if you never typed the word. You have probably seen this when search results seem to know what you meant.

I still type like I’m arguing with autocorrect, and Google figures out what I mean. Different studies call this by different names, and I do not claim a single definition. You can learn more about entities in another article I wrote.

Knowledge Graph vs. Knowledge Panel

Diagram contrasting the Knowledge Graph, a network of related entities, with the Knowledge Panel box Google shows when it is confident.

Here is a quick definition of each:

  • The Knowledge Graph is Google’s machine-readable encyclopedia of entities and how they relate.
  • A Knowledge Panel is the box in search results that summarizes a subject when Google is confident enough to show one.

If the panel appears for your brand, that is a good sign. If it does not, the main goal remains getting Google to understand you correctly.

Schema is Google’s native language

A student reviews a page of highlighted study notes beside a textbook, an analogy for schema giving Google a cheat sheet.

Before you panic, schema markup is not wizardry. Think of it as handing Google a cheat sheet for your page. You wrote a normal page for humans. Schema then tells Google, in structured data, “This page is about X. X is the same as this Wikipedia entry. Here are the key attributes.” It is supporting evidence, like giving a student study notes after a boring textbook.

Schema helps Google lock in what you mean, and it applies beyond rich snippets. I have seen it help pages get understood without keyword stuffing.

How Google detects entities

Flow diagram of Google breaking a page into many entities and reconciling them with known facts, rewarding consistent information.

Google reads your page with natural language processing. It uses the surrounding words to decide whether “engine” means a car part or a search engine. It breaks a page into hundreds of potential entities, then reconciles them with what it already knows. If your information around the web is messy or contradictory, Google’s “plate” looks cracked. Your job is to hand it a finished plate: neat, consistent, and obvious.

The three-step playbook I use

Infographic of a three-step playbook: pick an Entity Home, write a clear description, and corroborate the facts across the web.

I keep this simple so I can remember it without a PowerPoint.

  1. Pick an Entity Home
    Choose one page as the official home for a person, a product, or a brand. Google can reconcile fragmented facts there. Treat it as the source of truth.
  2. Write a clear description
    Explain it as if you were talking to a smart 12-year-old: who it is, what it is, what it does, and who it is for. Add schema that repeats the critical facts. Keep it tight.
  3. Corroborate across the web
    Make sure your facts match everywhere that matters: company listings, profiles, data sources, industry directories, and government registries. If the web repeats the facts you wrote, Google’s confidence increases. When your grocery bill looks like a car payment, you want your SEO to pay you back with certainty.

Topics vs. entities

Concentric diagram placing a core entity at the center, with related entities and topical authority built outward and joined by internal links.

Many of you are probably asking, “Aren’t topics and entities basically the same thing?” For most practical work, they are close enough. This is how I run it day to day:

  • Treat the brand, person, or product as the core entity.
  • Build topical authority around it with tightly related entities.
  • Link content internally so the relationships are obvious.

I work in this order: define the entity cleanly, then stack the topic layers on top. You can work in the opposite order and meet in the middle. Either way, connect the dots.

Moving from keywords to entities without breaking your brain

Hands assemble a banana split with bananas, ice cream, chocolate sauce, and whipped cream, showing supporting entities around one main topic.

Here is a transition plan that keeps your old spreadsheets:

  • Stop counting keywords. Identify what the page is about.
  • List the supporting entities. If the page is about banana splits, you will probably mention bananas, ice cream, bowls, chocolate sauce, whipped cream, and recipes. You do not need to say “bowl” ten times. Show the right context.
  • Use schema to mark the main thing. Tell Google what the page is primarily about, then reference the canonical identifiers for clarity.
  • Link internally with intent. Connect related pages so Google can follow the logic. The goal is a site-level knowledge graph where each page’s role is obvious.
  • Expect pleasant surprises. When you define entities well, Google can rank you for queries you did not target word-for-word, because the meaning matches. That is a sign your entity work is paying off.

How to become an authority on a topic

A speaker presents to a small attentive audience, showing authority earned by being recognized and cited by peers.

Authority takes work. Here is the recipe I use:

  • Publish content that answers real questions with correct detail and labels opinions as opinions.
  • Cover the topic’s core entities thoroughly, then expand into related subtopics.
  • Get recognized by peers and cited in places that matter. Third-party corroboration strengthens your authority.
  • Keep your facts consistent everywhere. If you change a title, date, or claim, update it across your whole ecosystem. I have a few regrets, and letting old bios rot is one of them.

Site-level knowledge graph in plain English

Network diagram of site pages, each about one primary entity and linked together, forming a site-level knowledge graph.

Schema plus internal links can form a knowledge graph of your site’s ecosystem. That is the goal. Make each page clearly “about” one primary entity, with sensible links to related entities. Done well, this makes your site clearer to Google.

Practical checklist you can use today

Infographic checklist of six entity SEO actions, from picking an Entity Home to linking related pages, ending with a bicycle metaphor.
  • Pick the Entity Home for your brand, products, and key people.
  • Write clear, child-proof descriptions and avoid jargon.
  • Add schema that matches the on-page facts.
  • Standardize facts across profiles and directories.
  • Map your pillar topics. Create or update content to cover the supporting entities.
  • Link pages to show the relationships between them.
  • Review your brand search result. If it is confusing, fix your ecosystem until it looks like your business card.

Keywords were the training wheels. Entities are the real bike. Get your facts straight, connect your ideas, speak Google’s language, and let the machine confidently match you with the right people. If you do that, you will build something that lasts.

I teach a course on entities and semantics in SEO called Treasure Map. You can also join my Facebook group, SEO Training Camp.