The entity-relationship model is fundamental to how language is processed and how search engines look at the page for relevancy. Instead of looking at word frequency; Content can be evaluated using related entities, and for the citations to closely related terms.
BERT (language model) is the processing model presented to the public in October 2019 by Google. It considers the context of how a word is used, based on what comes before and after the word, or named entity.
Processing introduces the need for additional cost and time, but the by product of the processing creates a entity dataset or knowledge panel. Pages not processed can be checked at scale against the dataset without the additional costs. The usage of the dataset may look like latent semantic indexing to many, the idea that Google was using LSI roughly appeared within the same time period; but the major difference is LSI uses a specific set of related words ... BERT determines the relationship between entities using algorithms, and can be more detailed.
Entity focused document retrieval is not a tenured technology. Its implementation will likely change over time with volatility driven by AI ... limited to the scope of semanticly related terms.
Search engines, and Google in particular, have many algorithms. Some people may think that these algorithms can actually read and understand the content. Then, on an emotional judgment, decide which content is best. Search engines actually just scan the content and sort pages based on mathematical algorithms.
An SEO Sile would be a group of pages related to a keyword with a hierarchical link structure to a central parent page. They provide easy navigation to all the sub-pages. Content silos can be stacked linking to additional silos.
Converting skyscrapers is akin to the avalanche SEO strategy; but the keywords or terms are based on the actual terms the site can, or should I say does, appear for in the SERPs. The site over a period of time continues to update and provide a gain of knowledge on the topic -- as an authoritative resource for the topic -- resulting in an upward trend in Google related to the subject matter for the site.
AI search optimization is optimizing for entity algorithms (Open AI, Google generative search, and voice search). These systems use the entity data to find facts and documents.
Entity-based AI search or (Open AI, Google generative search, and voice search), systems use the entity (relationship model) data to find fact and documents. Using schema is an SEO best practice, as search engines can use natural language which is a by-product of AI.
A brand has an identifiable relationship model with all related entities services or products it offers. When a brand comes into existence it is a clean slate, which can freely identify its products without the history or baggage an already existing term has.
Schema helps search engines by unambigously specifing entities with markup. The information helps search engines interpret content more accurately and may provides information that can be used for the Knowledge Panels, rich snippets, image SEO, and in other areas of Google.
Entities for SEO are the subject, topic or keyword of the page. They are a product of large language models, created and used by artificial intelligence.
An entity or the topic/keyword of a page can be a concept, idea, person, brand, place, thing, thing created by a person, or thing owned by a person. The entity data set contains popular and relevant sub-entities or facts. Some subjects have a breadth of popular sub-entities and others have very few sub-topics or facts of interest for the intent of search.
Search engines can use this entity data to determine relevancy and rank pages and sites. In Entity-Based SEO the best practice is to expand or define the relationship between entities. Optimization is achieved by both the frequency of usage of the entity on the page and the usage of semantic words and sub-entities on the page and page links.
The knowledge graph dataset include the topics or entities that are related to the subject of the search keyword, which can be used while indexing pages without the need for AI to examine the page, and usage of sub-entities can be quantified.
Entities Help SEO
For the usage of entities, consider entity based search is changing, and the bot is not actually reading the page. Bert is looking for entities in standard locations -- Looking back at Eliza in normal dialog the entity is at the begining of sentences, page title and headlines.
Entity-based search is a scalable improvement over using keyword density, and keyword proximity, which for SEO improves the exposure of topical silos, and pages relevant to the content people want to find. Keyword density and proximity are not dead but in many cases or for keywords where their is a breadth of information ... related entities can be used by search to find content which explains the subject better than content that uses the keyword more times.
In normal communications, a person first reveals or presents the main subject and then provides facts, questions, or details about it.
In some cases, the information may be about a detail. The main subject would be the main entity, the details would be its related sub-entities, and information about a detail would be a sub-entity of the sub-entity.
1st example: Pizza near me.
The main entity is pizza, and the related entity is a neighborhood busines. The requested information intent is the sub-entity "address" of the business. The "near me" provides search intent.
2nd Example: Pizza gluten free
Again the main entity is pizza, there is a related entity, and again the search intent is for information on the sub-entity of the sub-entity.
3rd Example: Pizza
The example does not have enough information to show intent. Information can be provided based on the most popular user intent related to the term, "near me."
Breadth-based results
Breadth refers to how wide or how many sub-topics a specific entity or word has. A search for tools has many types of tools ... Google will try to clarify which type of tool is being searched for by providing a "types of" search panel when searching for that word. A search for Vienna sausage does not have sub-types -- depth relevance becomes more important than breadth ... Google will provide a shopping panel.
Entity-based SEO is building on the breadth, which wants the relationship between the query and related entities. When breadth is not useful for a search query, depth remains relevant. But, it is important to note these are distinctly different content types and content strategies based on completely different algorithms.
Bottom of the sales funnel when somebody is looking to purchase a product ... content is depth-based. Top of the sales funnel or "awareness" when somebody is window shopping ... content is breadth-based.
Keywords vs. Entities
It is generally understood by Tenured SEO professionals that search engines improve and sometimes revolutionize how they retrieve and rank documents. A tenured SEO professional, knowing that algorithms change, adapts to new algorithms.
Keywords can be ambiguous and may reference more than one subject. An entity, on the other hand, references a single subject. Language translation using entities instead of keyword-by-keyword, provides better results. Entity-based algorithms within search makes voice search, chatbot, and AI search possible. Using the breadth of the entity the algorithm is able to find a matching document better than trying to find a document with both keywords.
Because entity-based algorithms are able to provide better results they are going to remain and have a greater effect on search and SEO into the future.
... Solution Smith tests SEO tactics so you don't have to ...
Full-stack SEO has many moving parts, and it takes time and effort to understand the nuances involved. Solution Smith takes care of the overhead costs of digital marketing, saving time and effort.
A search algorithms is based off of mathmatics and ranking factors must be quantifiable or countable ... factors that are only true or false can be used for quality indicators to determine if a site will be listed or buried.
