For a long time, the standard content strategy advice was to build keyword landing pages — one page per target keyword, optimised for the specific search term, structured to rank for that query. It is the approach that built most of the SEO playbooks that worked through the 2010s and into the early 2020s.
It is no longer the approach that works best. In AI search, where systems retrieve content through entity-driven fan-out rather than keyword matching, entity hubs outperform keyword landing pages on almost every dimension that matters — retrievability, citation likelihood, durability over time, and structural authority. This post explains the difference, why the shift matters now, and how to start building entity hubs without abandoning everything that already exists on your site.
What a keyword landing page actually is
A keyword landing page is a page built around a specific query. The target keyword might be something like best CRM for small business or email marketing automation tool. The page is structured to satisfy that query — meta title with the keyword, H1 with the keyword, body content that answers the query and incorporates related search terms, internal links from supporting articles pointing back.
This approach worked because traditional search ranking was largely keyword-driven. If your page was the best match for a query and had enough authority signals, it ranked. The page existed to capture a specific search demand.
The problem is that keyword landing pages are query-shaped. They are built around how a user might phrase a search at one specific moment. Search behaviour evolves; the page does not. AI fan-out generates dozens of sub-queries around any given topic; the keyword landing page addresses one of them well and the others poorly.
What an entity hub is instead
An entity hub is a page built around a specific entity — a product, a person, a concept, a methodology, an organisation — rather than around a specific query. The page exists to be the definitive source for that entity. It covers what the entity is, what its key attributes are, how it relates to other entities, what its history is, what its specifications or properties are.
The structural difference is important. A keyword landing page is shaped by the question. An entity hub is shaped by the thing being described. The entity hub answers many possible questions about that entity, including questions that have not been asked in that exact phrasing yet.
Some concrete examples of entity hubs:
- A page about your CEO that establishes them as the Person entity — biography, role, publications, speaking history, social profiles
- A page about a proprietary methodology that defines it as a Concept entity — what it is, how it works, what problems it solves, what its components are
- A page about a flagship product that defines it as a Product entity — features, use cases, integrations, pricing, target audience
- A page about your headquarters that defines it as a Location entity — address, regions served, history at the location
Each of these is a hub. Each one anchors an entity that the rest of your content can reference.
Why entity hubs outperform keyword landing pages in AI search
Five specific reasons the shift matters.
Fan-out coverage. AI systems generate sub-queries around entities. An entity hub addresses many sub-queries by virtue of comprehensively defining the entity. A keyword landing page addresses the one query it was built for and misses the rest of the fan-out.
Citation specificity. AI systems are more likely to cite content that clearly attributes facts to a specific entity. Entity hubs are structurally attributive — the entity is named, defined, and described in the page itself. Keyword landing pages often talk around the entity rather than naming it explicitly, which makes them harder to cite by attribution.
Schema reinforcement. Entity hubs are built to support schema markup that explicitly declares the entity type. Person hubs use Person schema. Product hubs use Product schema. Concept hubs use Article schema with about pointing to a concept. The schema and the content reinforce each other. Keyword landing pages often use generic WebPage schema because the page is not really about an entity in the same sense.
Durability over time. Search behaviour shifts. Keyword volume rises and falls. Specific phrasing changes. The keyword that mattered in 2022 may not be the keyword that matters in 2026. Entity hubs are more durable because the underlying entity does not change as fast as the search vocabulary around it. Acme Project Manager is still Acme Project Manager regardless of whether the dominant query phrasing is “best project management software” or “AI-powered project management” or whatever the language shifts to next.
Authority accumulation. Entity hubs can connect to each other through structured data relationships and internal linking, building an entity graph that compounds over time. Every new piece of content that references the CEO links to the CEO hub. Every product mention links to the product hub. Every methodology reference links to the methodology hub. The hubs accumulate authority as the site grows. Keyword landing pages typically do not accumulate this way because they are not designed to be referenced — they are designed to be ranked.
The transition strategy
You do not need to abandon your existing keyword landing pages. The transition strategy is additive — build entity hubs alongside the existing pages, then progressively rewire the internal linking and structured data to anchor the entity hubs as the canonical sources.
Step 1: Map the entities that should have hubs. From an entity blueprint, identify the Tier 1 and Tier 2 entities that most need a dedicated hub. Five to ten entities is usually the right scope to start. Skip Tier 3 entities — they do not yet justify dedicated hubs.
Step 2: Build the hubs for each entity. Each hub follows a similar structure:
– Strong H1 with the entity name
– Opening paragraph that defines the entity and establishes its primary attributes
– Sections covering the entity’s key attributes, history, relationships, and context
– Internal links to related entities in your site
– Schema markup declaring the entity type with @id
– sameAs references to authoritative external profiles where applicable
The length depends on the entity. A Person hub for a senior leader might be 800-1,500 words. A flagship Product hub might be 2,000-3,000 words. A Concept hub for a methodology might be 1,500-2,500 words.
Step 3: Rewire internal linking. Every existing page that mentions one of these entities should link to its hub. This is where most of the transitional work happens. A blog post that mentions Acme Project Manager should link to the Acme Project Manager hub. A case study that quotes the CEO should link to the CEO’s Person hub.
Step 4: Add structured data references. Pages that reference an entity should also reference it in their schema. An Article that quotes the CEO should declare the CEO as the author (or as referenced through mentions) using the CEO’s @id. The structured data graph and the content graph reinforce each other.
Step 5: Maintain. As new entities emerge (new products, new team members, new methodologies), new hubs should be created and the internal linking pattern should incorporate them. The transition does not have a final state — entity hub creation becomes an ongoing capability rather than a one-time project.
When keyword landing pages still make sense
There are still legitimate uses for keyword landing pages. A few cases where the older approach still works.
- High-intent commercial pages. A “request a demo” page, a pricing page, a comparison page targeting a specific high-intent commercial query. These pages are designed to convert traffic from specific queries, and the keyword framing serves that conversion goal directly.
- Programmatic pages with strong entity foundations underneath. Programmatic pages that follow EAV patterns (entity-attribute-variable) can effectively be both keyword landing pages and entity hubs simultaneously, because the entity structure is baked into the page generation.
- Short-term tactical campaigns. Pages built for a specific marketing campaign with a defined start and end date. These are not meant to compound — they are meant to capture a moment.
For most strategic content, though, the entity hub model is the more durable choice. The keyword landing pages that do still make sense should sit alongside entity hubs, not replace them.
The compounding effect
The reason entity hubs ultimately win is that they compound. Every hub you build adds to a graph that gets more valuable as it grows. The relationships between hubs strengthen each individual hub. The internal linking that points to each hub builds its authority. The schema markup that references each hub via @id connects it to the rest of the entity network.
Keyword landing pages do not compound this way. Each one is largely standalone. Building more keyword pages does not necessarily strengthen the existing ones. The work scales linearly: more pages, more traffic, but not more authority per page.
Entity hubs scale exponentially. Each new hub strengthens the hubs it relates to. Each new piece of content that links to existing hubs reinforces their authority. The graph effect is real and observable in citation patterns over time.
This is why the shift from keyword landing pages to entity hubs is not just an SEO tactic. It is a fundamental change in how content strategy generates durable value. The teams that make this shift early are the ones whose entity authority compounds while their competitors are still optimising for individual queries.
Continue your learning (MLforSEO)
This post covered the case for entity hubs and the transition strategy from keyword landing pages. The full implementation — including the entity hub templates for each entity type (Organisation, Person, Product, Concept, Location), the internal linking patterns that compound authority across hubs, the schema markup approaches that connect hubs into a traversable graph, and the BRIDGE framework that organises the full content-to-entity-graph workflow — is in the AI Search & LLMs: Entity SEO and Knowledge Graph Strategies for Brands course on MLforSEO.
Enrolling also gets you into the dedicated course channel inside the MLforSEO Slack community, where Beatrice Gamba and Lazarina Stoy answer course-specific questions and discuss ongoing implementation projects with course-takers.
Beatrice Gamba
The future of search and content discovery will be dialogical, personalized and agent-mediated. Digital leaders need to start integrating these concepts in their strategies to be ready for what’s coming.
- Beatrice Gamba
- Beatrice Gamba
- Beatrice Gamba
- Beatrice Gamba
