Wikidata, Wikipedia, and Knowledge Graph entity engineering This article serves as a comprehensive guide for establishing and managing business entities within Google's Knowledge Graph, utilizing Wikidata and Wikipedia as primary tools. It explains that entity recognition leads to Knowledge Panels, improved AI rankings, and enhanced brand visibility, while emphasizing that entity authority compounds over time through consistent references. The document is structured as both an installation manual and an audit reference, offering three operational modes—Install, Audit, and Hybrid—to build or evaluate entity recognition from scratch. Originally published atPart of ThatDevPro's open SEO + AI framework library. thatdevpro.com . ThatDevPro is an SDVOSB-certified veteran-owned web + AI engineering studio. Open-source AI citation toolkit: github.com/Janady13/aio-surfaces . Establishing Entities in Google's Knowledge Graph — Wikidata, Wikipedia, and the Entity Web A comprehensive installation and audit reference for establishing a business and its key entities in Google's Knowledge Graph, claiming and managing the Knowledge Panel, building reciprocal entity reconciliation across the web, and maintaining the entity authority that compounds across rankings, AI engines, and brand search results. This document is dual-purpose: installation manual and audit document. Cross-stack implementation note: the code samples in this framework are written in plain HTML for clarity. For React, Vue, Svelte, Next.js, Nuxt, SvelteKit, Astro, Hugo, 11ty, Remix, WordPress, Shopify, and Webflow equivalents of every pattern below, see . For pure client-rendered SPAs no SSR/SSG see framework-cross-stack-implementation.md . For Tailwind-specific concerns purge, dynamic classes, dark-mode CLS, focus accessibility see framework-react.md . framework-tailwind.md 1. Document Purpose & How to Use This Document 1.1 What This Document Is This is the canonical reference for establishing entities in Google's Knowledge Graph KG . The Knowledge Graph is Google's database of entities and their relationships — people, places, organizations, products, concepts, events. When Google recognizes an entity in its Knowledge Graph, that recognition manifests as Knowledge Panels in search results, populates AI Overviews, drives entity-based ranking, and feeds the data that LLMs train on. This framework specifies how to assess current entity recognition status, how to build the foundation for entity recognition Wikidata being the primary path , how to pursue Wikipedia inclusion when notability supports it, how to claim and optimize the Knowledge Panel when it appears, and how to maintain entity authority over time. It also specifies entity reconciliation — the process by which Google associates external profiles, references, and data points with the canonical entity in its Knowledge Graph. Entity authority compounds. Every authoritative reference, every Wikidata property added, every Wikipedia mention, every reciprocal external link reinforces Google's confidence in the entity. Conversely, weak or fragmented entity signals leave Google uncertain — and uncertain entities don't get Knowledge Panels, don't rank as authoritatively, and don't get cited by AI engines. 1.2 Three Operating Modes Mode A — Install Mode : Building entity recognition from scratch or adding to existing presence. Follow Sections 2 → 14. Mode B — Audit Mode : Evaluating existing Knowledge Graph status. Skip to Section 11. Mode C — Hybrid Mode : Audit then install for gaps. 1.3 How Claude Code CLI Should Consume This Document - Read Section 2 — collect client variables, especially existing entity references - Read Section 3 — understand the Knowledge Graph and entity ecosystem - Run Section 4 — assess current entity recognition status - Install Sections 5-9 — Wikidata first, then external reconciliation, then Wikipedia if notable, then Knowledge Panel claim - Validate — Section 11 - Generate report — Section 14 1.4 Conflict Resolution Rules | Conflict | Rule | |---|---| | Existing Wikidata entry with errors | Audit; correct via standard Wikidata editing process. Do not delete and recreate. | | Existing Wikipedia article with errors | Use talk page to discuss; let community editors handle changes. Don't edit your own entry. | | Multiple Wikidata entries for same entity | Use merge process; flag for review by Wikidata community. | | Knowledge Panel showing wrong information | Suggest edit through Knowledge Panel suggest-edit feature. | | Entity name conflict with existing notable entity | Disambiguate clearly; never try to claim someone else's entity. | 1.5 Required Tools - Wikidata — wikidata.org — primary platform for entity creation and management - Wikipedia — wikipedia.org — for notable entities - Google Search Console — to verify Knowledge Panel claim eligibility - Google Knowledge Graph Search API — developers.google.com/knowledge-graph — to verify entity recognition - OpenRefine — openrefine.org — for batch entity reconciliation - Wikidata Query Service — query.wikidata.org — for testing Wikidata entries 2. Client Variables Intake ============================================ KNOWLEDGE GRAPH FRAMEWORK CLIENT VARIABLES ============================================ --- Business Entity REQUIRED --- business name: "" Canonical name as entity business alternate names: All variations business legal name: "" If different from canonical business type: "" Schema.org type business founded year: "" business founder names: business founder qids: Wikidata QIDs if exist business industry: "" business industry qid: "" Wikidata industry classification business headquarters city: "" business headquarters country: "" business official website: "" --- Existing Entity Recognition Status REQUIRED --- business in knowledge graph: false Use Knowledge Graph Search API to check business knowledge panel appears: false Search business name; does panel appear? business wikidata qid: "" Existing Wikidata entry if any business wikipedia article exists: false business wikipedia article url: "" --- Founder Entity REQUIRED --- founder full name: "" founder alternate names: founder birth year: "" Optional but helps disambiguation founder credentials: founder employer qid: "" Wikidata QID for current/founded business founder in knowledge graph: false founder wikidata qid: "" founder wikipedia article exists: false --- Notability Assessment REQUIRED for Wikipedia --- business independent secondary sources: URLs of substantial coverage in independent publications business passes wikipedia notability test: false founder independent secondary sources: founder passes wikipedia notability test: false --- External Profile Inventory REQUIRED for sameAs reconciliation --- business linkedin url: "" business x url: "" business facebook url: "" business youtube url: "" business github org url: "" business crunchbase url: "" business bbb url: "" business chamber of commerce url: "" business industry directory urls: founder linkedin url: "" founder x url: "" founder github url: "" founder huggingface url: "" founder orcid: "" founder personal site url: "" founder other profile urls: --- Topical Entity Coverage RECOMMENDED --- primary topical entities: Topics covered with topical hub pages topical entity qids: {} Existing Wikidata QIDs for major topics --- Knowledge Graph Strategy Status REQUIRED --- has strategy for kg inclusion: false plans to pursue wikipedia inclusion: "" "yes now", "yes when notable", "no", "uncertain" plans to create wikidata entries: false has kg inclusion timeline: false --- Entity Maintenance REQUIRED ongoing --- wikidata last reviewed: "" wikipedia last monitored: "" knowledge panel last audited: "" external profile last audited: "" 3. What the Knowledge Graph Is Google's Knowledge Graph is a database of entities and their relationships. Launched in 2012, it powers: - Knowledge Panels in search results the boxes on the right side showing entity information - Featured snippets and rich results for entity-related queries - AI Overview source selection — AI Overviews preferentially cite Knowledge Graph entities - Voice search and assistant responses "Hey Google, who is..." - Entity-based ranking — pages associated with recognized entities benefit - Disambiguation — when multiple entities share names, KG provides the disambiguation infrastructure The Knowledge Graph contains hundreds of billions of facts about millions of entities. Most is automatically extracted from sources like Wikipedia, Wikidata, government databases, structured data on websites, and news mining. Some is manually curated. Entity inclusion in the Knowledge Graph isn't binary — it's confidence-based. Google may have very high confidence in well-established entities Apple Inc, Albert Einstein, Tokyo and lower confidence in less-established entities. Confidence determines whether Knowledge Panels appear, how prominently the entity features in results, and how authoritatively the entity is treated in AI Overviews. Entity confidence comes from: 1. Wikidata presence — Wikidata entries are direct input into Google's Knowledge Graph. A complete, well-cited Wikidata entry is the strongest single signal for entity inclusion. 2. Wikipedia presence — Wikipedia articles flow into Wikidata and are direct sources for Knowledge Panels. Wikipedia coverage is a strong signal but has higher notability requirements than Wikidata. 3. Schema markup with sameAs — Schema on the entity's official website that declares sameAs links to authoritative external profiles helps Google reconcile the entity. 4. Citation by other authoritative sources — When other recognized entities reference this entity, confidence increases. 5. Consistent factual data across sources — When the same facts appear across multiple authoritative sources, confidence increases. 6. Structured data accessibility — Sites that expose entity data via schema, API, or other structured formats are easier to ingest into KG. The 2026 evolution of the Knowledge Graph is significant. AI engines ChatGPT, Perplexity, Claude, Gemini all rely on entity data overlapping with Google's KG. Entities recognized in KG get cited more frequently across AI engines. The Knowledge Graph is no longer just for Google search — it's foundational infrastructure for the entire AI search ecosystem. For any business or individual seeking to be recognized as an authoritative entity in 2026, Knowledge Graph inclusion is foundational — not optional. 4. Current Entity Recognition Assessment Before pursuing inclusion, understand current status. 4.1 Knowledge Graph Search API Check Use Google's Knowledge Graph Search API to check if entities are already recognized: curl "https://kgsearch.googleapis.com/v1/entities:search?query={{ENTITY NAME}}&key={{API KEY}}&limit=10" The response shows whether Google has the entity, its type, description, and confidence score. If the entity returns: - Strong match resultScore 100 : Entity is well-recognized - Weak match resultScore <100 : Some recognition but not strong - No match : No KG presence For this site's primary entities business, founder, key topics , document current status. 4.2 Knowledge Panel Search Test In incognito mode, search for the entity name and document: - Does a Knowledge Panel appear on the right side? - What information is in the panel? - Is the information accurate? - Is the panel "claimed" has owner verification ? - What sources does the panel cite Wikipedia, official website, etc. ? Test variations: {{exact business name}} {{business name}} {{location}} {{business name}} {{founder name}} {{founder full name}} who is {{founder name}} 4.3 Wikidata Direct Search Search Wikidata directly: wikidata.org/w/index.php?search={{ENTITY NAME}} Document: - Existing entries that match - Existing entries that conflict other entities with same name - Whether existing entry is complete and accurate 4.4 Wikipedia Search Search Wikipedia: en.wikipedia.org/w/index.php?search={{ENTITY NAME}} Document: - Existing article if any - Quality and completeness of article - Whether the article links to the official website 4.5 External Profile Mapping Document all existing external profiles for the entity. Each profile is a potential sameAs reconciliation point: business external profiles: - platform: "LinkedIn" url: "" verified: false matches official data: true - platform: "X / Twitter" url: "" verified: false matches official data: true - platform: "Crunchbase" url: "" matches official data: true ... etc Identify any profiles with mismatched data different addresses, different founding dates, etc. . These need correction before reconciliation, or they'll undermine entity confidence. 4.6 Recognition Status Summary After assessment, classify status: - Established entity — Strong KG presence, accurate Knowledge Panel, complete Wikidata, possible Wikipedia. Focus on maintenance and refinement. - Recognized but partial — Some KG presence, partial info, no Knowledge Panel or unclaimed panel. Focus on completion and claiming. - Minimal recognition — Some external profiles but no KG presence. Focus on creating Wikidata entry and reconciliation. - No recognition — No external profiles, no Wikidata, no KG. Foundation must be built from scratch. The implementation path varies by status. 5. Wikidata Implementation Foundation Wikidata is the primary path to Knowledge Graph inclusion. Wikidata has a lower notability bar than Wikipedia and is direct input to Google's Knowledge Graph. Every entity that wants KG inclusion should have a Wikidata entry as foundation. 5.1 Wikidata Notability Requirements To create a Wikidata entry, the entity must satisfy at least one notability criterion: - Has a Wikipedia article in any language automatic notability - Refers to an instance of a clearly identifiable conceptual or material entity most businesses, people, places qualify - Fulfills a structural need to make statements about other entities For most businesses and individuals associated with businesses, criterion 2 applies. Notability bar is essentially: is this a real, identifiable entity that people might reasonably want to make statements about? 5.2 Wikidata Account Setup Create a Wikidata account at wikidata.org . Use a real account with established edit history if possible — accounts with no history can have edits scrutinized more carefully. Best practice: declare conflict of interest on user page if creating entry for own business or self. 5.3 Creating the Business Entity Navigate to wikidata.org/wiki/Special:NewItem . Required initial fields: Label — {{business name}} canonical form, capitalized correctly Description — Concise under 250 char , defining sentence. Examples: - "American technology company headquartered in Cupertino, California" Apple Inc. - "Service-Disabled Veteran-Owned web development and SEO firm based in Cassville, Missouri" ThatDeveloperGuy Aliases — All alternate names. List every legitimate alternate spelling, abbreviation, DBA name. After creating the basic entry, populate properties. 5.4 Required Wikidata Properties for Businesses Add these properties Wikidata uses property IDs like P31, P17 — find each on Wikidata's property documentation : | Property | Property ID | Value Source | |---|---|---| | Instance of | P31 | Type — "business" Q4830453 , "company" Q783794 , or more specific | | Country | P17 | Country US = Q30 | | Headquarters location | P159 | City Wikidata QID | | Founder | P112 | Person Wikidata QID create if needed | | Inception | P571 | Founding date | | Industry | P452 | Industry QID | | Official website | P856 | Domain URL | | Logo image | P154 | Wikimedia Commons file upload first if not there | | Coordinates | P625 | Lat/long for physical location | Example of a populated business entry: Item: Q138610626 ThatDeveloperGuy Label en : ThatDeveloperGuy Description en : Service-Disabled Veteran-Owned web development and SEO firm based in Cassville, Missouri Statements: - instance of P31 : business Q4830453 - country P17 : United States of America Q30 - headquarters location P159 : Cassville Q... - founder P112 : Joseph Anady Q... if created - inception P571 : 2020 - industry P452 : web development Q386275 - official website P856 : https://thatdeveloperguy.com - coordinates P625 : 36.6770° N, 93.8730° W External Identifiers: - LinkedIn P4264 : {{linkedin handle}} - Crunchbase P2087 : {{crunchbase id}} 5.5 Required References for Each Property Every statement should be supported by a reference: inception P571 : 2020 References: - stated in: official website P856 of subject - reference URL: https://thatdeveloperguy.com/about/ - retrieved: 2026-04-29 Without references, statements can be challenged. With strong references, the entry is robust against editor revisions. 5.6 Creating the Founder Person Entity Same process for founder. Required Person properties: | Property | Property ID | Value | |---|---|---| | Instance of | P31 | human Q5 | | Country of citizenship | P27 | Country | | Date of birth | P569 | If notable; can omit for privacy | | Place of birth | P19 | If notable | | Occupation | P106 | Occupation QIDs | | Employer | P108 | Business QID the entity created above | | Educated at | P69 | School QIDs | | Field of work | P101 | Field QIDs | | Native language | P103 | Language QID | | Official website | P856 | Personal site if exists | External Identifier properties for founder: | Identifier | Property ID | |---|---| | LinkedIn personal profile | P6634 | | GitHub username | P2037 | | Hugging Face username | P9100 | | ORCID iD | P496 | | X Twitter username | P2002 | 5.7 Topical Entity Wikidata Entries For unique topical entities the site is the authority on specific frameworks, methods, tools the site has created , create Wikidata entries. Most general topics already have entries. For example, if the site has authored a specific framework called "SDVOSB Engine Optimization Methodology," that could be a Wikidata entity: Item: Q... SDVOSB Engine Optimization Methodology Label en : SDVOSB Engine Optimization Methodology Description en : A 14-tier framework for search engine and AI engine optimization developed by Joseph Anady at ThatDeveloperGuy Statements: - instance of P31 : methodology Q1799072 - developed by P178 : Joseph Anady Q... - subclass of P279 : search engine optimization Q180711 - depicts P180 : 14-tier framework 5.8 Cross-Linking via External Identifiers Wikidata's external identifier properties allow direct linking to authoritative external databases. Add as many as apply: For businesses: - LinkedIn company ID P4264 - Crunchbase organization ID P2087 - BBB business profile ... if property exists - Industry-specific identifiers BSI, NAICS code, etc. For people: - LinkedIn personal ID P6634 - ORCID P496 - Google Scholar author ID P1960 - ResearchGate profile P6178 - GitHub username P2037 External identifiers strengthen entity reconciliation — they tell systems unambiguously that this Wikidata entity is the same entity as in those external sources. 5.9 Wikimedia Commons Logo Upload Upload the business logo to Wikimedia Commons: - Visit commons.wikimedia.org/wiki/Special:UploadWizard - Confirm copyright logo must be licensed permissively or be your own work you license - Upload with descriptive filename: ThatDeveloperGuy-logo.svg - Add to Wikidata entry as logo image P154 The logo is then available for use in Knowledge Panels and other systems. 5.10 Patience for Wikidata Propagation After creating Wikidata entries, allow time for propagation: - 24-72 hours for community visibility - 1-4 weeks for Google's Knowledge Graph ingestion - 1-3 months for stable Knowledge Panel appearance typically requires multiple signals beyond Wikidata Don't expect immediate Knowledge Panel from Wikidata alone. Wikidata is foundational; other signals build on it. 6. External Reconciliation Strategy Beyond Wikidata, build the network of references that reinforce entity recognition. 6.1 Schema sameAs on Official Website The official website's Organization and Person schema must include sameAs links to all authoritative external profiles AND to the Wikidata entry: