Google does not understand the web the way a human does. It does not read a website and “understand” what that company does. It builds a graph of entities, nodes connected by relationships, and places each brand within this graph based on the signals it finds. If those signals are missing, inconsistent or contradictory, the brand remains an ambiguous entity: partially understood, inaccurately described, rarely cited by AI systems.
Entity SEO and the Knowledge Graph are not technicalities for specialists: they are how you build a digital identity that search engines and AI systems recognise, understand and respect.
Ai/© 2026
LLM
Knowledge Graph
Google doesn't read pages.
It reads entities.
Is yours defined?
The foundation
Google's Knowledge Graph:
a universe of connected entities
The Knowledge Graph is the structured database Google uses to understand the real world: not web pages, but the entities that web pages describe. People, companies, places, products, events, concepts: each entity is a node in the graph. The nodes are connected by relationships: “Bliss Agency” is based in “Rome”, operates in the “communications and marketing” sector, was founded in “year X”, works with “luxury sector clients”.
When Google displays the Knowledge Panel, the side box with information about a brand in the SERPs, it is reading from the Knowledge Graph. When Gemini generates an answer about a company, it draws on the same source. When ChatGPT with web browsing looks up information about a brand, the structured sources it consults (Wikipedia, Wikidata, Crunchbase) are the same ones that feed Google’s Knowledge Graph.
The practical consequence is simple: a brand that is well defined in the Knowledge Graph is better understood, described more accurately and cited more frequently by every system, Google Search, Google Gemini, ChatGPT, Claude, Perplexity. A brand that is absent or poorly defined is read ambiguously, described with errors and ignored in generative answers.
01.
Headquarters and location
Where the entity operates in the physical world. It connects the brand to the graph’s geographical entities: Rome, Italy, Europe.
02.
Sector and category
The semantic territory in which the brand operates. It determines which category queries the brand is considered relevant for.
03.
Products and services
What the entity offers. Connects the brand to product/service entities and the associated transactional queries.
04.
Associated people
Founders, CEO, team. Connects the brand to person entities, strengthening E-E-A-T and sector authority.
05.
Authoritative sources
Where the Knowledge Graph finds reliable information about the entity: Wikipedia, Wikidata, editorial publications.
06.
Related entities
Competitors, partners, cited clients. The network of relationships defines the entity’s competitive positioning in the graph.
The approach
Entity SEO: optimising
identity, not just pages
Traditional SEO optimises pages: keywords, meta titles, backlinks, speed. Entity SEO optimises identity: who the brand is, what it does, where it operates, who it is associated with, how it is described in authoritative sources. They are not mutually exclusive; they complement each other. But Entity SEO adds a dimension that traditional SEO alone does not cover.
The central concept is that of a semantically stable entity: a brand that Google always recognises as the same entity, regardless of how it is talked about. If it is called “Bliss Agency” on the website, “Bliss S.r.l.” on LinkedIn, “Bliss Agency Roma” on Wikipedia and “Bliss, Agenzia di Comunicazione” on Google Business, Google struggles to link the four versions into a single coherent profile. The Knowledge Graph accumulates uncertainty instead of authority.
Identity consistency is the prerequisite for everything else. Before any advanced semantic optimisation, the brand must have a structured, consistent and verifiable identity across all the sources Google consults.
500B+
entities in Google’s Knowledge Graph, people, places, companies, products, concepts
E-E-A-T
the Google evaluation framework that the Knowledge Graph feeds directly
17
Verified sameAs URLs in Bliss Agency’s Organization schema implementation
sameAs
The brand's identity network:
the sameAs field in the Organization schema
The sameAs in the Organization schema is an array of URLs that tell Google where to find authoritative information about the brand entity on other platforms. It is the mechanism by which the brand declares its distributed identity: “these profiles all refer to us”. Google uses these declarations to aggregate information from all sources into a coherent profile of the entity in the Knowledge Graph.
A sameAs that is incomplete means Google cannot connect the scattered information about the brand and treats it as separate or ambiguous entities. A sameAs field, when complete, creates a network of cross-confirmations that strengthens the entity’s identity in the graph, increasing E-E-A-T authority and the likelihood of citation in AI systems.
Sources to include in Organization sameAs - Priority
Wikipedia - Critical
The structured source with the greatest weight in the Knowledge Graph. A verified, consistent Wikipedia entry is the strongest authority signal for defining the entity.
Wikidata - Critical
The structured database linked to Wikipedia that Google uses directly to feed the Knowledge Graph. It allows the entity's attributes to be defined in machine-readable format.
Google Business Profile - Critical
The verified profile on Google, a direct signal to Google itself about the brand's identity. It verifies the match between the web entity and the real-world entity.
LinkedIn Company - High
The professional platform with the greatest authority in the graph for corporate entities. Consistency of name, sector and description between LinkedIn and the website is essential.
Crunchbase - High
A company database used by ChatGPT, Claude and Gemini as a source of information on companies and start-ups. Essential for B2B and tech brands.
Official social profiles - High
Instagram, X/Twitter, Facebook, YouTube: all the brand's official social profiles must be included to complete the distributed identity network.
Business Register / ATECO - Medium
For Italian companies, verified presence in the Registro delle Imprese (Companies Register) is a legitimacy signal that strengthens E-E-A-T Trust.
Sector directories - Medium
Recognised sector directories (Clutch for agencies, Booking for hotels, TripAdvisor for restaurants) specific to the entity's sector.
The technical language
Organization schema:
how to declare the entity to Google
Organization schema markup in JSON-LD format is how a brand communicates its structured identity directly to Google and AI systems. It is not a technical optional: it is the formal declaration of the entity. A website without a properly completed Organization schema forces Google to infer information about the brand from external sources, with all the risks of inaccuracy that follow.
// Implementazione schema Organization completa
{
"@context": "https://schema.org",
"@type": "Organization",
"name": "Bliss Agency",
"legalName": "Bliss S.r.l.", // ← nome legale coerente
"url": "https://blissagency.it/en/marketing-communications-brand-advisory-agency/",
"logo": "https://blissagency.it/logo.svg",
"description": "Agenzia di comunicazione e SEO...",
"foundingDate": "2019",
"address": {
"@type": "PostalAddress",
"streetAddress": "Via del Mandrione, 63",
"addressLocality": "Roma",
"addressCountry": "IT"
},
"knowsAbout": ["SEO", "GEO", "Branding", "Digital PR"],
"areaServed": "Italia",
"sameAs": [ ← il campo più importante
"https://it.wikipedia.org/wiki/Bliss_Agency",
"https://www.wikidata.org/wiki/QXXXXXXX",
"https://www.linkedin.com/company/bliss-agency",
"https://www.crunchbase.com/organization/bliss",
"https://www.instagram.com/blissagency.it",
// ... 12 ulteriori URL verificati
]
}
Each field of the Organization schema carries a specific weight for the Knowledge Graph. The sameAs, name, legalName and url fields are the most critical, as they determine the entity’s unique identity. The knowsAbout, areaServed and description fields define its semantic territory. Bliss checks every URL in the sameAs before including it: a sameAs pointing to an outdated or inconsistent profile produces contradictory signals.
External sources
The platforms that feed the Knowledge Graph
The Knowledge Graph is not built from the website alone: it is built from an ecosystem of external sources that Google consults to triangulate information about the entity. Each source adds a layer of confirmation to the brand’s identity, and consistency across sources is the strongest reliability signal a brand can send to the Knowledge Graph.
Wikipedia and Wikidata
The sources that carry the most weight in the Knowledge Graph. A coherent, well-structured Wikipedia entry is the strongest signal of entity authority. Wikidata adds the machine-readable structure that Google uses directly to feed the graph.
Impact: Critical · used by all AI models
Google Business Profile
Verifying the brand on Google Business Profile is the most direct signal that an entity is real and operational. It confirms the match between the web entity and the entity in the physical world, which is essential for E-E-A-T Trust.
Impact: Critical · generates Knowledge Panel
Crunchbase and LinkedIn
Company databases used by ChatGPT and Perplexity as primary sources of information on companies. Consistency of name, sector, headquarters and description across Crunchbase, LinkedIn and the website is the minimum requirement for a company entity.
Impact: High · essential for B2B
Consistent editorial mentions
Every press article that mentions the brand contributes authority signals and semantic associations (sector, expertise, positioning) to the Knowledge Graph. Consistency of information across mentions is critical: contradictions between sources create ambiguity in the entity.
Impact: High · E-E-A-T Authoritativeness
Verified social profiles
Instagram, LinkedIn and X/Twitter with verification badges strengthen the entity’s legitimacy. Social profiles do not feed the Knowledge Graph as directly as Wikipedia, but they contribute to the sameAs network and to the consistency of the distributed identity.
Impact: Medium · identity consistency
Schema markup on the website
The Organization JSON-LD schema is the only source the brand controls directly. It declares the brand’s identity to Google in a structured way and provides the sameAs links to all other sources. It is the starting point of Entity SEO — but it is not sufficient on its own without external confirmation.
Impact: Fundamental · starting point
the method
Four levels of coverage.
One continuous system.
Positioning on Google is not a finish line; it is a condition to be maintained. Bliss governs its clients’ organic positioning through four operational levels that work in parallel and reinforce one another.
How does Google see the brand now: coherent or ambiguous?
The starting point is mapping the current entity: searching for the existing Knowledge Panel, checking the consistency of information across the website, Google Business Profile, Wikipedia, LinkedIn and the main structured sources, analysing the existing sameAs in the Organization schema, and identifying the inconsistencies that create ambiguity in the Knowledge Graph. Guiding question: if Google had to describe this brand in three sentences, what would it say? And would it be correct? The Entity Audit answers these questions before any intervention.
Declaring identity unambiguously. Field by field.
Bliss implements the Organization schema with all relevant fields: name and legalName consistent with all external sources, a description optimised for target sector queries, knowsAbout with the key concepts of the positioning, a precise areaServed, foundingDate, and verified contact details. The sameAs field is built with all verified URLs — each URL is checked manually before inclusion to ensure the information is consistent and current. On Bliss itself: 17 verified sameAs URLs.
The sources the Knowledge Graph considers most authoritative.
For brands that meet Wikipedia's notability criteria (editorial coverage in recognised publications, documented sector relevance), Bliss supports the creation or improvement of the Wikipedia entry — respecting the platform's editorial guidelines and ensuring that the information is accurate, verifiable and consistent with the website and other sources. The Wikidata entry is created or updated with all the attributes relevant to the company entity: industry, country, founders, products, main competitors.
Name, Address, Phone: the same everywhere. Always.
NAP (Name, Address, Phone) consistency across every source where the brand appears — website, Google Business Profile, LinkedIn, Crunchbase, Wikipedia, sector directories — is the fundamental requirement for the entity's stability in the Knowledge Graph. Bliss maps every occurrence of the brand on the web, identifies inconsistencies and resolves them systematically. This work has an immediate impact on both Local SEO and the quality of the entity in the Knowledge Graph.
The Knowledge Graph changes. The entity must be maintained.
The Knowledge Graph is not static: it updates according to the new information Google finds on the web. A change of headquarters, a new product, a partnership with a well-known brand, a significant editorial mention — all these events alter the entity's profile in the graph. Bliss monitors changes to the Knowledge Panel monthly, checks sameAs consistency after every change in external sources and updates the Organization schema promptly. The entity is treated as an asset to be maintained, not a one-off intervention to be forgotten.
THE TEAM
Our voices
leading the way
Bliss’s SEO is managed by a team with experience of organisations with complex sales cycles in tech, manufacturing and professional services: sectors where the buyer is a decision maker with structured evaluation processes and long purchasing timelines.
Andrea Maso
HEAD OF SEO
International background, a data-driven mind and no intention of slowing down.
Mario Antonini
chief design officer (cdo)
Luca Maletta
Head of Copywriting
Suraksha Kumari
Full stack developer
Develops SEO-oriented architectures and optimises performance, Core Web Vitals and technical structure.
Sajjad Mazaherizaveh
Full stack developer
Combines front-end and back-end development with technical SEO and AI Search Optimisation.
FAQ
Frequently asked questions about the Knowledge Graph
What is Google's Knowledge Graph?
The Knowledge Graph is the structured database of entities — people, companies, places, products — and their relationships. Google uses it to understand who the subjects discussed on the web are, how they are connected and which information is reliable. The Knowledge Panel in the SERPs and Gemini’s answers both draw on the Knowledge Graph.
Building your presence in the Knowledge Graph means existing as a recognisable, well-defined entity for Google and for AI systems — not as a collection of web pages, but as a real-world entity with precise attributes.
What is Entity SEO?
Entity SEO optimises how a brand is represented as an entity in the Knowledge Graph. It works not on individual web pages but on the brand’s structured identity: who it is, where it operates, what it does and who it is connected with. It differs from traditional SEO because it works on identity, not on keywords.
An entity that is well defined in the Knowledge Graph is understood more precisely by Google, cited more accurately by AI systems and associated with the right sector queries — with effects on both organic ranking and GEO visibility.
How is a presence in the Knowledge Graph built?
Through: complete Organization schema markup with sameAs pointing to Wikipedia, Wikidata and official profiles, a verified presence on Google Business Profile, consistent Crunchbase and LinkedIn profiles, editorial backlinks from authoritative sources, and NAP (Name, Address, Phone) consistency across all sources.
The guiding principle is consistency: the same information must appear identically across every source Google consults. Contradictions between sources create ambiguity that weakens the entity in the graph.
What is sameAs and why does it matter?
The sameAs field in the Organization schema is an array of URLs that tell Google where to find authoritative information about the entity on other platforms: Wikipedia, Wikidata, LinkedIn, Crunchbase, social profiles, Google Business Profile.
sameAs allows the Knowledge Graph to aggregate information from all these sources into a coherent profile. An incomplete sameAs means Google treats the brand’s different presences as separate or ambiguous entities — dispersing authority instead of accumulating it.
Why does the Knowledge Graph matter for visibility in AI systems?
Gemini draws directly on Google’s Knowledge Graph to build its answers about brands. ChatGPT and Claude use the same structured sources (Wikipedia, Wikidata, Crunchbase) that feed the Knowledge Graph. A brand that is well defined in the graph is described more accurately and cited more frequently by all AI systems.
Knowledge Graph optimisation is the foundation of GEO strategy — before any content optimisation or LLM Digital PR, the entity must be defined consistently and recognisably.
How do you obtain a Google Knowledge Panel?
The Knowledge Panel is generated automatically by Google when the entity is sufficiently well defined in the Knowledge Graph. It is not requested: it is built. The prerequisites are: a complete Organization schema with sameAs pointing to Wikipedia and Wikidata, a verified Google Business Profile, and editorial backlinks from authoritative sources.
Once generated, the Knowledge Panel can be claimed by the brand owner through Google to verify its accuracy and add official information.
What is the difference between the Knowledge Graph, the Knowledge Panel and schema markup?
The Knowledge Graph is Google’s internal database of all entities and their relationships. The Knowledge Panel is the visual presentation of an entity in the SERPs: the “shop window” of the Knowledge Graph. Schema markup (Organization in JSON-LD) is the language we use to communicate information about the entity to Google from our website.
Schema markup feeds the Knowledge Graph, which in turn generates the Knowledge Panel. All three are necessary: schema without Wikipedia is not enough; Wikipedia without schema loses precision.