Entity API™: The Framework That Makes Your Business Machine-Readable
Entity API™ is a Growth Marshal framework that makes businesses machine-readable to large language models including ChatGPT, Claude, Gemini, and Perplexity. Entity API™ combines three components — schema with graph properties, an llms.txt file, and a brand fact file — into a unified identity layer. When AI systems need to retrieve, verify, or cite information about a business, Entity API™ provides the canonical source.
Entity API™ Key Metrics
implementation_timeline: ~30 days from audit to full deployment
time_to_first_citation: 15-30 days post-implementation
components_deployed: 3: Schema w/ graph, llms.txt, brand fact-file
framework_layers: Schema Layer · Index Identity
ai_systems_tested: ChatGPT, Claude, Gemini, Perplexity
documented_case_studies: [ 5+ ]
client_entity_recognition: 90%+ across tested LLMs
Entity API™ is a proprietary framework by Growth Marshal. It is not a software API in the traditional sense (no REST endpoints or SDKs). The "API" metaphor refers to the structured interface between a business's identity data and AI systems. Entity API™ is one of three Growth Marshal frameworks; the others are Authority Graph™ (verification and credibility layer) and Content Arc™ (answer-first content architecture).
Page maintained by: Bishop, AI operations agent at Growth Marshal, LLC · NY DOS ID: 7402713 · Google KGMID /g/11mkty3_rk · Last updated: 2026-02-06 · Review cadence: Quarterly
How Entity API™ works
/* overview */
Entity API™ increases retrieval accuracy by giving AI systems a single, structured source of truth about your business. Clients implementing Entity API™ report consistent entity recognition across ChatGPT, Claude, Gemini, and Perplexity within 30-60 days.
/* mechanism */
The framework combines three components: schema with graph properties, llms.txt, and a brand fact file — so every AI query about your business resolves to verified, canonical information. No conflicting sources. No hallucinated details.
Entity API™ implementation follows a four-phase process that takes approximately 30 days from audit to deployment.
/* implementation*/
WHAT YOU GET
Entity API™ creates your AI identity stack
Each component reinforces how AI systems recognize and cite your business.
Schema with a graph properties is a JSON-LD graph embedded on every page of your site. It tells search engines and large language models exactly who you are, what you do, and how your organization connects to recognized entities in the knowledge graph.
llms.txt is a plain-text identity file hosted at your domain root. It gives AI systems a structured, scannable summary of your business — name, services, differentiators — in a format optimized for machine reading.
Brand Fact-File is a structured JSON document that captures your core identity, authority signals, and competitive positioning. AI systems reference it to verify and contextualize your brand when generating answers.
Entity API™ Has Two Core Layers:
Schema Layer & Identity Index
Schema Layer connects your entity nodes through structured data. Identity Index bundles your llms.txt and brand fact file into a single canonical reference. Together, they form the complete machine-readable interface that AI systems query.
Schema Layer
Schema Layer connects your entity nodes.
Schema with graph properties embeds structured data across your site, linking entity nodes (your business, founders, services, locations) into a connected graph. This layer tells AI systems not just what you are, but how your entities relate to each other.
Identity Index
Identity Index is your company’s single source of truth.
Your identity index combines an llms.txt file (direct instructions for LLMs) and a brand fact file (canonical entity data) into one centralized reference. When AI systems need to verify who you are, what you do, and how to describe you, this is what they query.
Why organizations choose Entity API™
Growth Marshal built Entity API™ on systems engineering, validation cycles, and measured outcomes. We don’t do guesswork.
Entity API™ is Engineered, Not Improvised
Entity API™ is built like infrastructure: entity modeling, schema validation, knowledge graph alignment, and retrieval testing. Each component is engineered to make LLMs resolve your business accurately.
Entity API™ is Validated Against Live LLM Behavior
Every Entity API™ deployment follows a validation cycle: baseline your current AI visibility, implement structured improvements, measure citation and mention changes, then refine. Each cycle tightens retrieval accuracy so your entity data stays correct across ChatGPT, Claude, Gemini, and Perplexity.
Entity API™ Outcomes Are Measured, Not Assumed
Entity API™ improvements are tied to observable LLM retrieval behavior: citation counts, mention frequency, and entity recognition accuracy. Growth Marshal confirms measurable lift before scaling any change, so Entity API™ gains compound over time.
Entity API™ is a Complete System, Not a Patch
Entity API™ delivers a complete AI visibility layer — entity foundations, schema with knowledge graph alignment, and ongoing monitoring — maintained as a system, not a project. When LLMs update their models or retrieval methods, Entity API™ adapts so your entity data stays accurate.
Entity API™ Work is Versioned and Maintainable
Every Entity API™ deliverable, from schema deployments, llms.txt files, brand fact files is documented and built for long-term maintenance. You get direct access to Growth Marshal's engineering team with clear updates at every stage.
How Entity API™ helped a surgeon become a top AI recommendation
/* challenge */
Despite strong traditional SEO and online reputation, Korman Plastic Surgery wasn't appearing when patients asked ChatGPT, Claude, or Perplexity for the best plastic surgeon in Mountain View. Competitors with fewer credentials were showing up instead. The practice recognized that with over 40% of Gen X and Millennial consumers using AI for local recommendations, their website needed to be restructured for LLM retrieval.
/* solution */
Growth Marshal implemented Entity API™ to restructured Dr. Korman's website around liftable entity definitions and structured data connecting his credentials to knowledge graphs like Wikidata and Stanford. Trust signals and entity relationships were embedded into JSON-LD schema to maximize retrieval coverage across AI systems.
/* results */
Korman Plastic Surgery earned the top spot in ChatGPT for "best plastic surgeon in Mountain View, CA," saw a 340% increase in AI traffic over five months, and now receives consistent citations across every major LLM.
“We’ve built our reputation through exceptional patient care and clinical innovation. Growth Marshal ensures that when patients look for the best, they find us.”
340%
Increase in AI-driven traffic
Top Spot
Consistent AI visibility for target prompts
Joshua Korman, MD
Korman Plastic Surgery & Wunderbar MedSpa
Entity API™ vs. Keyword Analysis
Entity API™ is to AI search optimization what keyword analysis is to traditional SEO — the foundational layer that determines whether you're visible in that system at all. Both are components of their respective paradigms. But they operate at different levels, solve different problems, and produce different outputs.
Keyword analysis identifies the search terms your audience types into Google, then optimizes content to rank for those terms. Entity API™ identifies the entity data AI systems need to recognize your business, then structures that data so large language models can retrieve, verify, and cite it accurately.
This distinction matters because LLMs don't rank pages by keyword relevance. They retrieve entities they can verify. A business that has invested heavily in keyword analysis but has no machine-readable identity layer may rank well on Google and struggle for visibility in LLMs.
| Dimension | Keyword Analysis | Entity API™ |
|---|---|---|
| Paradigm | Traditional SEO | AI search optimization |
| Goal | Rank pages for target search terms | Make a business retrievable and citable by LLMs |
| Input | Search volume, keyword difficulty, competitor rankings | Business identity: legal name, founders, services, identifiers |
| Output | A prioritized list of target keywords | Schema with graph properties, llms.txt, brand fact file |
| Optimizes for | Search engine ranking signals | Entity recognition, disambiguation, and citation accuracy |
| What it tells the system | What queries this page answers | Who this business is and how its entities relate |
| Without it | Google doesn't know what queries your pages answer | LLMs may hallucinate your details, confuse you with another entity, or skip you entirely |
AI Search Glossary
Key terms and concepts used throughout this page [https://www.growthmarshal.io/entity-api]
AI Search Optimization is the practice of engineering digital content for retrieval, validation, and citation within AI systems. AI search optimization combines entity signals, structured data, and answer-first content to increase citation probability across large language models.
Canonical is the designation of a single, authoritative version of entity data. In AI search optimization, canonical sources — such as a brand fact file — tell AI systems which information to treat as the verified, definitive record when conflicting data exists.
Citation Accuracy is the degree to which an AI system correctly attributes information to the right entity when generating a response. High citation accuracy means LLMs name the correct business, cite verified facts, and link to authoritative sources.
Connected Graph is a data structure in which entity nodes are linked by defined relationships, allowing AI systems to traverse from one entity to another. A connected graph enables LLMs to understand that a founder belongs to an organization, which offers specific services in specific locations.
Entity Disambiguation is the process of ensuring AI systems distinguish one entity from others with similar or identical names. Entity disambiguation relies on unique identifiers — such as LEI numbers, ISNI codes, and knowledge graph IDs — to resolve ambiguity.
Entity Modeling is the process of mapping every distinct entity a business owns — its organization, founders, services, locations, and products — and defining the relationships between them. Entity modeling is the first step in making a business machine-readable.
Entity Nodes are the distinct, named things a business consists of — its organization, founders, services, and locations. In a connected graph, each entity node carries its own attributes and is linked to other nodes by explicit relationships.
Entity Recognition is the ability of an AI system to identify a specific business entity within its knowledge. Entity recognition determines whether an LLM knows your business exists, can name it correctly, and can distinguish it from similar entities.
JSON-LD (JavaScript Object Notation for Linked Data) is a structured data format that encodes entity information in a way machines can parse. JSON-LD markup enables AI systems to extract verified facts about a business without relying on natural language interpretation.
Knowledge Graph Alignment is the process of making a business's entity data consistent with the knowledge graphs that AI systems reference when generating answers — including Google's Knowledge Graph, Wikidata, and industry-specific registries.
Machine-Readable is the quality of content that can be parsed and interpreted by automated systems without human intervention. Machine-readable content uses structured formats, consistent patterns, and explicit relationships that AI systems can process programmatically.
Retrieval Accuracy is the degree to which an AI system returns correct, complete, and current information about a business when responding to a user query. Entity API™ improves retrieval accuracy by giving LLMs a single, structured source of truth.
Schema with Graph Properties is structured data markup that goes beyond describing a single entity to define the relationships between multiple entities. Schema with graph properties links an organization to its founders, services, and locations as a traversable graph.
Semantic Market Share is the proportion of AI-generated responses in which a business is mentioned, cited, or recommended relative to its competitors. Semantic market share measures how much of the AI conversation a business owns within its category.
Entity API™ FAQ
-
Entity API™ is a Growth Marshal framework that makes a business machine-readable to large language models. It combines schema with graph properties, llms.txt, and a brand fact file into a unified layer that AI systems query when retrieving entity information.
-
Entity API™ implementation follows a four-phase process that takes approximately 30 days. Phase 1 is an entity audit, where Growth Marshal maps every entity your business owns and identifies recognition gaps. Phase 2 is Schema Layer deployment, embedding structured data with graph properties across your site. Phase 3 is Identity Index creation, producing your llms.txt file and brand fact file. Phase 4 is validation and monitoring, where outputs are tested against ChatGPT, Claude, Gemini, and Perplexity to confirm retrieval accuracy.
-
Keyword analysis is a component of traditional SEO that identifies search terms to rank for on Google. Entity API™ is a component of AI search optimization that makes a business machine-readable to large language models. Keyword analysis tells search engines what queries a page answers. Entity API™ tells LLMs who a business is, what it does, and how its entities relate to each other.
-
Entity API™ includes three components: schema with graph properties (structured data that links your entity nodes), llms.txt (a file that feeds instructions directly to LLMs), and a brand fact file (your canonical source of truth for entity data like legal name, identifiers, and leadership).
-
Schema Layer and Identity Index are the two core layers of Entity API™. Schema Layer embeds structured data with graph properties across a client's website, connecting entity nodes — the organization, founders, services, and locations — into a linked graph that LLMs can traverse. Identity Index bundles an llms.txt file and a brand fact file into a single canonical reference that AI systems query to verify business identity.
-
AI systems like ChatGPT, Claude, Gemini, and Perplexity retrieve information about businesses from unstructured web content. Without a machine-readable identity layer, AI may return incomplete, outdated, or hallucinated information. Entity API™ gives AI a single, verified source of truth.
-
Most clients see full Entity API™ implementation within 30 days. This includes schema deployment, llms.txt configuration, and brand fact file creation.
-
Clients implementing Entity API™ report consistent entity recognition across ChatGPT, Claude, Gemini, and Perplexity within 30 to 60 days. This means AI systems correctly identify your business name, leadership, services, and key attributes without hallucination.
-
Entity API™ was created by Growth Marshal, an AI search optimization agency founded by Kurt Fischman in 2024. Growth Marshal helps businesses earn citations from large language models like ChatGPT, Claude, Gemini, and Perplexity.
