What Entity Extraction Is, and Where It Fits in SEO

Entity extraction is, to my mind, one of the most underused techniques in SEO, and one of the most important to understand, because it is close to how machines actually read text. Search engines and large language models do not see your content as a bag of keywords. They see entities, the real-world things your content is about, and the relationships between them. Learn to see text the way they do and you can build much better strategies. This post covers what entity extraction is, where it sits in the wider world of language understanding, how entities differ from keywords, the EAV model, and a proper tour of where all of it fits in SEO.

What entity extraction (NER) is

Entity extraction, also called named entity recognition or entity chunking, is a natural language processing technique that identifies and classifies key pieces of text, the named entities, into predefined categories like people, locations, organisations, events and dates. It takes text written for humans and adds structure and semantic information a machine can use, which is why it is such a common preprocessing step, including inside the large language models generating answers today.

In machine learning terms, NER is supervised and prediction-based: it is trained on labelled examples, then predicts whether the entities it finds in new text match the ones locked in its knowledge graph, using context, surrounding words and other entities to decide. It can run on any kind of text, social posts, legal documents, scientific papers, and any length, from a query to a full page.

Where entity extraction sits: semantic analysis

It helps to know the neighbourhood NER lives in, because it explains why the same APIs do so many related jobs.

Entity extraction sits within semantic analysis, alongside sentiment analysis and topic modeling

Entity analysis is one part of semantic analysis, the phase of language understanding concerned with the meaning of text, which also includes sentiment analysis and topic modelling. This is why most pre-trained NLP APIs (Google Cloud’s, Amazon’s, IBM’s) bundle entity extraction together with sentiment, entity sentiment, content moderation and classification. The practical upshot: besides a list of entities, you get entity type, salience (importance), sentiment, mention counts and the different phrasings used, plus metadata like a Wikipedia URL or a Google Knowledge Graph ID. That extra data is what turns “here are some entities” into real analysis.

Entities are not keywords

This distinction trips up a lot of SEOs, so it is worth nailing.

Keywords are text properties we value for SEO reasons; entities are distinct, well-defined concepts that exist in the real world. Take the word “cone”, it could mean three different things depending on context, and only entity recognition resolves that. In a query like “shop online Nike Air Force One”, an SEO reads transactional and navigational intent from the words, while entity extraction identifies Nike (organisation), Air Force One (product) and Michael Jordan (person). Sometimes keywords and entities overlap; often they do not. Niche terms like “SEO research” get recognised as entities but classified as “other” by general pre-trained models, simply because they are not prominent in the training data.

The EAV model: entities, attributes, variables

Entities get much more powerful once you add structure to them.

The entity, attribute, variable (EAV) model with a worked example

Every entity type has attributes (properties), and every attribute has variables (specific values). Take “dog” as an entity: an attribute is “breed”, and its variables are Labrador, Husky and so on. This EAV model is the backbone of programmatic SEO. Take “influencer” as an entity, add attributes like location and family status, and combine variables: “influencers in London”, “influencers married with kids in the beauty niche”. Each combination is a tailored landing-page target. It is also central to information-gain analysis and competitor content analysis. (We go deep on EAV in the Semantic Keyword Research course.)

How entity extraction models are built

If you ever consider training your own, there are four approaches, and knowing their trade-offs tells you why you probably should not start there.

Four approaches to building entity extraction models: rule-based, classic ML, deep learning, hybrid

Rule-based is easy to implement and precise for clear patterns, but brittle with ambiguity and language variation. Classic machine learning trains a classifier (like an SVM) on labelled examples. Deep learning and hybrid architectures, which combine segmentation, tokenisation, rule-based, lexicon-based and deep-learning components, are what the big providers use, because they have the data and engineering to make them perform. You can custom-train or fine-tune with tools like spaCy, adding contextual embeddings, conditional random fields and active learning. But entity extraction is genuinely labour-intensive to build well, so for beginners a pre-trained API like Google Cloud NLP or Amazon Comprehend gets you to a result far faster. Only self-train if you have unique domain entities, strict data-privacy needs, or a real desire to learn the pipeline, and even then, fine-tuning usually beats training from scratch.

Where entity extraction fits in SEO

Once you can extract entities at scale, a lot opens up. And the data you can feed it is everywhere: long-form text (your content, competitor content, YouTube and podcast transcripts, Reddit, PDFs, patents), short-form text (comments, social posts, brand mentions), metadata (titles, URLs, meta descriptions, anchor text) and queries (ranked and user queries from Google, YouTube or TikTok).

SEO uses of entity extraction: keyword research, programmatic SEO, SERP analysis, internal linking, content audits, topic maps

Keyword research gets an entity layer: validate which keywords contain entities, map closely linked entities into content, build lists of entity attributes and variables for your writers, and check which entities sit in the knowledge graph. Programmatic SEO practically requires it, EAV combinations at scale become your page structure. SERP analysis is a big one: scrape titles, meta descriptions and URLs for your branded queries, extract the entities, and see which entities appear, how often, and how their sentiment relates to ranking position (with the eternal caveat that correlation is not causation).

Internal linking improves when you link articles by shared entities and tune anchor text to entity mentions. Content audits use entity prominence as a lens: flag an entity as prominent if it appears more than a set number of times per page, then check whether your best page on that entity actually covers its attributes, whether the entity appears in the title and meta, and whether other articles link to it with the entity in the anchor text, while spotting thin articles with no entities at all for consolidation or sunsetting.

And topic maps, the modern way to plan content, are built by grouping entities into clusters, combining entity analysis, EAV query exploration, SERP analysis and your content audit into a genuinely comprehensive map of what to produce to stand out.

Entity extraction also powers entity-driven feedback and review analysis, mapping how positively customers talk about specific aspects of your service, and digital PR tracking, measuring the sentiment your brand is mentioned with across ranked coverage.

Try it free

Run your first extraction with the free Entity Extraction with Google’s Natural Language API guide, no code, entities in seconds.

Where this goes next

This post covers what entity extraction is and where it fits. The Entity Extraction and Analysis module inside Introduction to Machine Learning for SEO then goes hands-on: comparing NLP APIs against generative AI, query entity extraction, semantic analysis of customer reviews, and entity-based internal linking. Read the companion post on which NLP API to actually use next.