What is Generative Engine Optimisation (GEO)?

The next evolution of ecommerce visibility in the AI era.

For more than two decades, ecommerce visibility has been dominated by search engines.

If a brand wanted customers to find its products online, it invested in Search Engine Optimisation (SEO): optimising pages to rank for keywords in search results.

But discovery behaviour is changing rapidly.

Consumers are increasingly turning to AI assistants such as ChatGPT, Gemini, and Perplexity to ask questions directly:

• What’s the best espresso machine for beginners?
• Which standing desk should I buy for a small office?
• What are the best lightweight backpacks for travel?

Instead of showing a list of links, these systems generate answers and recommendations.

Data already shows how quickly this behaviour is emerging. AI-referred retail traffic grew 35x between July 2024 and May 2025, according to Adobe Analytics research on ecommerce referral patterns.

This shift is creating a new optimisation discipline: Generative Engine Optimisation (GEO).

The shift from search engines to answer engines

Traditional search engines return ranked lists of pages.

Users must evaluate those pages themselves.

The classic SEO-driven journey looks like this:

Keyword -> Search engine -> List of links -> User comparison -> Purchase

AI assistants change this dynamic.

Instead of presenting links, they analyse information and generate a direct recommendation.

The discovery journey increasingly looks like this:

Question -> AI analysis -> Synthesised answer -> Product recommendation -> Click -> Purchase

In this model, the AI assistant becomes a decision layer between consumers and brands.

And the competitive surface is much smaller.

AI systems typically surface only a small shortlist of products per query. If a product does not appear in that shortlist, it may never be considered by the customer at all.

The key question for ecommerce brands is therefore no longer simply:

Can customers find my pages?

Instead it becomes:

Can AI systems understand and recommend my products?

What is Generative Engine Optimisation?

Generative Engine Optimisation (GEO) is the practice of structuring product information so that AI systems can interpret, evaluate, and recommend it.

Where SEO focuses on ranking pages, GEO focuses on making products intelligible to AI systems.

The term gained wider attention following a 2024 academic study by researchers from Princeton University, Georgia Tech, and IIT Delhi. Their research showed that structured, clearly attributed content increased visibility in generative AI responses by up to 40% compared with unstructured alternatives.

At a practical level, GEO involves three core layers.

1. Discovery signals

AI systems still rely on technical signals to locate and interpret content.

Important discovery infrastructure includes:

• XML sitemaps
• internal linking
• structured metadata
• emerging standards such as llms.txt

These signals help AI crawlers identify what content exists on a site.

Cloudflare Radar research shows that AI-driven crawling activity increased more than 15x in 2025, as AI assistants increasingly fetch live information to answer user questions.

If a page cannot be discovered by these crawlers, it cannot be recommended.

2. Semantic understanding

Large language models interpret content differently from traditional search engines.

Instead of focusing primarily on keyword matching, they attempt to understand:

• product attributes
• relationships between variants
• category structures
• suitability for specific use cases

Structured formats such as JSON-LD schema and product attribute data help AI systems interpret these relationships.

For example, a product description that says “premium quality running shoe” provides little machine-readable context.

But structured attributes such as:

• cushioning level
• stability type
• terrain compatibility
• weight
• price range

allow an AI system to reason about whether the product fits a particular query.

3. Intent alignment

Consumers rarely ask AI assistants for specific SKUs.

Instead, they ask intent-driven questions such as:

• Best travel backpacks for carry-on luggage
• Running shoes for flat feet
• Standing desks under £500

To answer these questions effectively, AI systems prefer content organised around clusters of products that match the intent.

This means discovery increasingly depends on intent-aligned product groupings, rather than isolated product pages.

Why traditional SEO is no longer enough

SEO remains important, and many principles of good SEO (clear structure, relevant content, technical accessibility) also support GEO.

But search engines and AI assistants operate differently.

Search engines:

• rank pages
• return links
• let the user compare options

AI assistants:

• interpret product attributes
• synthesise information
• generate recommendations

This difference changes the optimisation problem.

A page that ranks well in traditional search results may still fail to appear in AI recommendations if the underlying product data lacks clarity or structure.

How AI assistants evaluate ecommerce content

Large language models build internal representations of information, often described as knowledge graphs.

These graphs map relationships between entities such as:

• products
• brands
• categories
• attributes
• use cases

A well-structured product might be represented conceptually like this:

Running Shoe
├ Cushioning -> Maximum
├ Stability -> High
├ Arch Support -> Structured
├ Terrain -> Road
├ Weight -> 280g
└ Price -> £120

When a user asks about running shoes for flat feet, the AI system can directly evaluate which products have attributes that match the requirement.

If those attributes are missing or poorly structured, the AI may not have enough signal to recommend the product.

GEO vs SEO

The relationship between SEO and GEO is best understood as an evolution rather than a replacement.

Both approaches will coexist.

However, as AI assistants become a larger part of product discovery, GEO will become an increasingly important layer of ecommerce visibility.

Why this matters for ecommerce brands now

AI-driven discovery is still early, but several signals suggest it is growing quickly.

• AI-referred ecommerce traffic has increased dramatically in recent years.
• AI crawlers are becoming significantly more active across the web.
• Structured product data improves the likelihood of appearing in AI responses.

For merchants, the implication is straightforward:

Products that are easier for AI systems to understand are more likely to be recommended.

The next era of ecommerce visibility

The history of ecommerce discovery has been shaped by technological shifts.

First came web directories.

Then search engines transformed how consumers found products.

Now AI assistants are becoming the next major interface between consumers and the internet.

In this new environment, success will depend not only on ranking pages but on ensuring products can be understood, evaluated, and recommended by AI systems.

Generative Engine Optimisation is the discipline that makes this possible.

Frequently asked questions

What is Generative Engine Optimisation (GEO)?

GEO is the practice of structuring product information so AI assistants such as ChatGPT, Gemini, and Perplexity can understand and recommend products.

Is GEO the same as SEO?

No. SEO focuses on ranking pages in search results. GEO focuses on making products understandable to AI systems so they can be recommended in answers.

Does GEO replace SEO?

No. Both will coexist. However, AI-mediated discovery is growing rapidly, which means product visibility increasingly depends on AI understanding.

What platforms does GEO apply to?

GEO principles apply to any ecommerce platform. Shopify and WordPress are currently the most common environments where structured GEO optimisation is being adopted.

How do I get started with GEO?

Start with a product data audit:

• Are product attributes structured clearly?
• Are relationships between products defined?
• Do discovery pages exist for common intent queries?

Platforms such as Geoffy help automate this process.

Sources

• Adobe Analytics: AI-driven retail traffic insights
  https://business.adobe.com
• Aggarwal et al. (2024), GEO: Generative Engine Optimization
  https://arxiv.org/abs/2311.09735
• Cloudflare Radar 2025: AI crawler activity
  https://blog.cloudflare.com/radar-2025-year-in-review

About the author

Anthony Gale is Co-Founder of Geoffy, a Generative Engine Optimisation platform for ecommerce brands.

He has spent more than two decades working in ecommerce and digital growth, helping retailers adapt to major shifts in online discovery.