1. Definition
Nesting JSON-LD for Depth is an Advanced Schema.org technique in Generative Engine Optimization (GEO) where multiple, related Entities are defined within a single JSON-LD block by referencing one entity from within the definition of another. This creates a clear, structured Entity Graph on the page. Instead of simply defining a single WebPage, nesting allows a brand to define a Review of a Product, written by a Person (Author), published by an Organization.
This depth signals a high level of Information Gain and Citation Trust to Large Language Models (LLMs), as it explicitly outlines complex factual relationships in a machine-readable format.
2. The Mechanics: Building the Entity Graph
Nesting mimics real-world relationships, which the LLM can then easily map to its internal Knowledge Graph. The core mechanism involves using an entity’s property value to contain the full definition of another entity.
Simple vs. Nested Schema
| Schema Type | Description | LLM Perception |
| Simple (Flat) | Defines one entity (e.g., a Product) and references the author’s name as a plain string. | Low confidence in the author’s E-E-A-T. |
| Nesting (Deep) | Defines the Product and uses the author property to contain the full definition of a Person entity (including their jobTitle, sameAs links, etc.). | High confidence; the LLM knows the author’s credentials and Entity Authority. |
The LLM Advantage
When the LLM encounters a nested structure, it can extract complex, verifiable facts in a single operation. For example, when asked, “Who is the expert who wrote the review for product X?”, the nested Schema provides a single, unambiguous answer: {Review Entity} $\rightarrow$ {Author Property} $\rightarrow$ {Person Entity} $\rightarrow$ {sameAs Link}. This structured clarity boosts the content’s Citation Trust Score.
3. Implementation: Technical Best Practices for GEO
Nesting is achieved by using the object (curly brackets {}) structure within the property value, rather than a simple string or URL.
Focus 1: Defining Authorship for E-E-A-T
Nesting the Person entity within the author or reviewedBy property is the most essential GEO application of nesting.
Code snippet
{
"@type": "Article",
"headline": "...",
"author": { // NESTED ENTITY STARTS HERE
"@type": "Person",
"name": "Jane Doe",
"jobTitle": "Lead Security Researcher",
"sameAs": [
"https://www.linkedin.com/in/janedoe"
]
}, // NESTED ENTITY ENDS HERE
"publisher": "..."
}
GEO Impact: The jobTitle and sameAs links within the nested Person definition provide explicit evidence of Expertise and Authoritativeness to the LLM.
Focus 2: Product Reviews and Offerings
For commercial content, nesting a Review or AggregateRating within a Product is crucial.
Code snippet
{
"@type": "Product",
"name": "Geo-Optimized Widget",
"review": { // NESTED ENTITY 1 (The Review)
"@type": "Review",
"reviewBody": "Excellent for GEO...",
"reviewRating": {
"@type": "Rating",
"ratingValue": "4.8"
}
},
"offers": { // NESTED ENTITY 2 (The Offer)
"@type": "Offer",
"priceCurrency": "USD",
"price": "99.99"
}
}
GEO Impact: This structure enables the LLM to instantly extract and synthesize specific facts (price, rating, reviews) for use in AI Overviews and comparison chips, maximizing the content’s Information Gain.
Focus 3: The @id and @type Tags
Nesting must be consistent. Every entity defined inside another must start with its own @type tag. Using the @id tag on the main entities allows different Schema blocks on the page to reference the same entity without redefining it repeatedly, which simplifies parsing for the LLM.
4. Relevance to Generative Engine Intelligence
By using nested Schema, a brand ensures that all facts are presented with the highest possible degree of context and verifiability. This is the key to winning complex, multi-faceted queries where the LLM must synthesize information about not just what the page is, but who wrote it and what they said about the product.