JSON-LD

Definition

JSON-LD (JavaScript Object Notation for Linked Data) is a method of encoding structured data using the familiar JSON format. It allows website owners to embed machine-readable information directly into their web pages, making it easy for search engines, AI agents, and other automated systems to understand what a page is about.

In ecommerce, JSON-LD is most commonly used to mark up product information - name, price, availability, reviews, images - using Schema.org vocabulary. When you see rich snippets in Google search results showing star ratings, prices, and stock status, that data typically comes from JSON-LD markup on the product page.

JSON-LD is embedded in a page’s HTML inside a <script type="application/ld+json"> tag. Unlike older structured data formats like Microdata or RDFa, which weave annotations into the HTML content itself, JSON-LD sits in a separate block. This makes it easier to implement and maintain because it does not require modifying the visible page structure.

Why It Matters

JSON-LD has been the foundation of structured product data for over a decade, and its importance is growing in the age of AI-driven commerce. Here is why:

Search engine visibility. Google explicitly recommends JSON-LD as its preferred structured data format. Product pages with proper JSON-LD markup are eligible for rich results - enhanced search listings that show price, availability, and reviews directly in the search results. These rich results consistently earn higher click-through rates.

AI agent comprehension. When AI agents crawl the web or process web pages, JSON-LD provides clean, structured data they can parse without guessing. An AI agent reading a product page with JSON-LD can instantly extract the exact price, SKU, availability status, and description - no web scraping heuristics needed.

Bridge to agentic commerce. While protocols like ACP and MCP represent the next generation of AI-commerce interfaces, JSON-LD remains the most widely deployed structured data format. For merchants who cannot yet implement MCP or ACP, well-implemented JSON-LD is the most accessible way to make products AI-readable today.

Platform support. Most major ecommerce platforms generate JSON-LD automatically. Shopify, WooCommerce, PrestaShop, and BigCommerce all include product JSON-LD in their default themes or through popular plugins. This means many merchants already have some JSON-LD in place, even if they have never heard the term.

How It Works

JSON-LD uses Schema.org vocabulary to describe entities and their properties. A basic product markup includes the product name, description, image URLs, price, currency, availability, and brand. Here is what the structure looks like conceptually:

The markup defines a Product type with properties like name, description, and image. Inside it, an Offer object specifies the price, currency, availability status, and the URL where the product can be purchased. A Brand object identifies the manufacturer. An AggregateRating object can summarize customer reviews.

For ecommerce, the most important Schema.org types to mark up with JSON-LD are:

• Product - the core product entity with name, description, SKU, and images
• Offer - pricing, availability, and purchase conditions
• Review / AggregateRating - customer feedback and ratings
• BreadcrumbList - category navigation path
• Organization - the merchant’s business information

Search engines and AI systems read these JSON-LD blocks when processing a page. Google uses them for rich results. AI agents like those powered by GPT or Claude use them to understand product data when browsing the web. The more complete and accurate your JSON-LD, the better these systems can represent your products.

The key for merchants is completeness. A JSON-LD block with just a product name and price is technically valid but far less useful than one that includes images, descriptions, variant information, reviews, brand, and availability. Every field you populate is another data point that search engines and AI agents can use to match your product to a shopper’s query.

Related Terms

• Schema.org - The vocabulary standard that JSON-LD uses to describe products and other entities
• Structured Data - The broader category of machine-readable data that JSON-LD implements
• Product Feed - Another mechanism for distributing structured product data to external systems
• Generative Engine Optimization (GEO) - Optimization practices where JSON-LD plays a foundational role