Artificial intelligence
Customer & Marketing
GEO and structured data
Publiée le January 7, 2026
Publiée le January 7, 2026
Structured data (schema.org) has long been a pillar of SEO. It enables search engines to understand the structure of content. In the context of GEO, the question arises: should we continue to use schema markup?
AIs rely on explicit signals to interpret and select passages. Boulder SEO Marketing points out that GEO “incorporates advanced schema markup techniques to help AI understand content structure”. Tags (FAQPage, HowTo, Article, LocalBusiness, Product) make it easier for the models to read and increase the chances of being cited.
JSON-LD (recommended by Google) offers a flexible, remote solution that doesn’t affect HTML structure. For GEO :
Article: includes title, description, publication date, author and E-E-A-T fields.
FAQPage: defines Q/Rs, improving visibility in extracts and reuse by AIs.
HowTo: useful for procedural content, it defines steps and a result.
Product: specifies features (name, description, price, availability) and may include notices.
LocalBusiness: describes a business with location information and attributes (services, opening hours).
| Type of diagram | Use case | GEO impact |
|---|---|---|
| Article | Blog articles, studies, guides | Allows AI to understand subject, date and author |
| FAQPage | FAQ sections | Improves likelihood of being quoted for specific questions |
| HowTo | Step-by-step tutorials | Enables structured extraction of steps |
| Product | Product sheets, comparisons | Provides structured details that AI can integrate into its responses |
| LocalBusiness | Store pages | Helps AI associate locations, services and opening hours |
Respect the hierarchy. Each schema type must be properly nested and conform to the schema.org standard.
Include UGC. Add “review” and “aggregateRating” fields with verified reviews.
Avoid spam. Use realistic, relevant data; artificial markup can be ignored or penalized.
Update data. Make sure your information (products, prices, opening hours) remains up to date.
Some marketers question the relevance of schema markup in the GEO era. In reality, schema remains a crowbar for organizing information and making it comprehensible to AIs. It acts as a lexicon describing entities and their relationships. Without a schema, the AI has to guess at the organization of your page; with one, it understands explicitly that it’s an article, tutorial or product sheet.
Structured data formats include Microdata, RDFa and JSON-LD. Google now recommends the use of JSON-LD, a JavaScript script that does not alter the HTML structure. This format is more flexible and easier to maintain. Microdata or RDFa embedded directly in HTML may be relevant for simple sites, but they complicate maintenance when you need to modify the structure.
As part of GEO, some publishers are adopting a llms.txt file (similar to robots.txt) to indicate which paths on their site can be used by LLM models. Although this practice is still experimental, it is intended to guide AIs in extracting the most relevant content. Oomph refers to this trend and recommends monitoring engine developments.
Structured data must be tailored to your products or services. Here are the most common types and their applications:
| Schematic type | Usage | Examples of key fields | GEO tips |
|---|---|---|---|
| Product | Product data sheets (e-commerce) | namedescription, price, availability, brand, review |
Include UGC reviews and availability information to improve the quality of generative responses |
| LocalBusiness | Store or agency pages | nameaddress, telephone, openingHours, geo, aggregateRating |
Add services, photos, local FAQs and synchronize with GMB |
| HowTo | Step-by-step guides | name, step, tool, image |
Break down each step into short sentences, add images and videos |
| Event | Event announcements (conferences, webinars) | namestartDate, location, performer, offers |
Clearly indicate location, time and price; refer to sites such as Eventbrite |
| Review and AggregateRating | Customer reviews | reviewBody, reviewRating, author, datePublished |
Include verified testimonials and add a ratingValue medium to build trust. |
| FAQPage | FAQ sections | mainEntity (list of Question/Answer) |
Cover several variations of questions and provide concise answers |
Suppose you’re selling headphones. You can structure the product sheet using the tags Product and Offer:
This precise structure enables AI to identify the brand, price, availability and reviews. By adding FAQPage markup at the bottom of the file (questions on autonomy, compatibility, warranty), you further improve readability.
To find out if your beaconing is bearing fruit :
Use Search Console to check that your rich snippets are detected and displayed in SERPs (zero position or product maps).
Monitor your GEO metrics: a well-structured schema can increase the likelihood of being cited by the AI and improve your ASoV. Test your pages in ChatGPT and Perplexity to see if the AI picks up your information correctly.
Compare two similar pages, one with schema and one without, and measure the difference in citations and conversions.
Beyond current practices, structured data will evolve to integrate the new needs of AI. Initiatives such as Schema for AI seek to enrich schemas to include information on data provenance, licensing or source quality. In the near future, AIs will be able to use this metadata to assess the reliability and diversity of content. For example, a field indicating that the statistic comes from an official body could be used as a signal of authority. It is also likely that engines will impose restrictions via llms.txt files or dedicated HTTP headers to control AI access to content. Publishers will need to keep pace with these developments and adapt their schema accordingly.
The schema markup remains a valuable tool for GEO. It makes it easier for models to understand your pages, and serves as a guarantee of reliability. In an environment where AIs extract and summarize information, providing a clear and complete schema increases your chances of being mentioned and helps users to trust your products and services.
Schema markup remains a major asset for GEO. By structuring your data, you enable AI engines to reliably understand and reuse your content, increasing your chances of citation and visibility.