Entity Based Searching

Are you restructuring your site to better answer questions from searchers? Here are our thoughts on entity based search, its value, and a quick trick to help establish your entity in the all-seeing eyes of Google.

In our last installment, we discussed ways to optimize your site for semantic search, without altering your site’s code or markup. To search semantically is simply to ask questions of a search engine using your natural language. And more than likely, you’re asking questions about a particular entity – a brand, an actor, a movie, an event.

The intent of Google’s Knowledge Graph was to give an identity to every entity in the world. Identity can include facts such as date of birth, important films, place of birth. So, when you ask Siri or Google Now voice assistants “Where was Barak Obama born?” – you get an answer. Not a list of sites where you can find the answer.

Google research has shown that on more difficult queries, people first search using their natural language questions. This means longer queries on average. But as the queries failed, the searches would revert to keyword queries.

So there is certainly value in Google developing a better understanding of natural search. But there is nothing we as SEOs can do to optimize for every single phrase or long tailed keyword.

[We know you try – we try too!] Predicting the questions and answers of searches isn’t exactly a science… yet.

So, how can you structure [or restructure] your site to help Google connect the dots between questions and answers? So that the content makes sense, and the relationship between the words and their context make sense? To predict a searcher’s intent? Because, what we’ve tried before with link building isn’t enough on its own, and never has been. Links have been manipulated over and over. Enter Entity Search, which gives you the chance to define what your site is all about.

Here is one quick idea to help Google figure out your entity:

You need to be using Schema.org markup for your sites.Search engines want machine-readable content to provide more accurate answers to searchers’ questions. Your markup teaches Google what entities are on your site. But that’s not the “quick idea” part.

Here’s the “quick idea” part. Take it one step further using the “sameAs” property in your markup to tell Google that the entities on your site as the same as entities on sites like Wikipedia or Freebase.com.

For example, let’s say your website is about the actor Tom Hanks. Clearly, there are multiple sites on the web about Tom Hanks. Multiple pages, and one entity. The “sameAs” relationship connects the dots between the pages and entity, helping to develop absolute identity. Using the “sameAs” property to tell Google that the Tom Hanks of your site is also the Tom Hanks of Wikipedia, Freebase, IMDB, etc. develops a much richer connection.

A little research into how to better define your entity will tell you that you also need to:

  • Use lots of nouns [preferably the noun that = your entity] in your writing
  • Keep linking to relevant sites about your entity
  • Optimize your Google+ listing. It’s appearing more and more in Knowledge Graph entries. Let’s talk more about this one later, shall we?
  • Update and edit your company’s Wikipedia and Freebase entries.

If you are already using Scheme.org on your site, utilizing the “sameAs” property should be a quick way to establish some authority for your site. There are very few quick fixes in our world of SEO… so take this one and run with it!

Published: January 11th, 2015 | Updated: April 11th, 2017|Tags: |0 Comments