Jonathan Mall

Generative Engine Optimization (GEO): How Your Brand Gets Recommended in AI Search in 2026

Viele Quellen, drei Empfehlungen — wie Generative Engine Optimization die KI-Auswahl bestimmt
Key takeaway: This GEO guide by AI keynote speaker Dr. Jonathan Mall shows how Generative Engine Optimization (GEO) decides which brand ChatGPT, Perplexity and Google AI Overviews recommend in 2026 — on average only 2.8 brands per query. Seven concrete strategies, a real AI-visibility scan of 293 brands, and an FAQ help you become the recommendation instead of staying invisible.

Why Google Alone Isn’t Enough Anymore: How Brands Become Visible in AI Search

There is a new reality in marketing, and most companies haven’t grasped it yet: your customers no longer ask Google. They ask ChatGPT, Perplexity, and Google AI Overviews.

And these AI systems don’t return ten blue links. According to current analyses, ChatGPT names on average only 2.8 brands per query. Either you’re one of those recommendations — or you’re invisible. A user who asks “What’s the best tool for audience analysis?” gets three names, no scrolling, no comparing, no page two.

That’s the core of Generative Engine Optimization (GEO): the practice of structuring content, entities, and data so generative AI systems select your brand as the answer. In my earlier article, “Recommended or Invisible” (German), I laid out the underlying problem. Here I show what you can actually do about it — including results from our AI Visibility Check (German case study), in which we systematically scanned 293 brands.

The pace of this shift is no longer a niche concern: ChatGPT now accounts for roughly 60.7% of all AI search traffic, Google AI Overviews appear on more than 25% of all searches, and 25% of B2B buyers already prefer AI-powered search over Google. Gartner predicts that by 2028, around 90% of all B2B purchasing processes will involve AI agents.

How AI Search Systems Choose Brands

Before you can optimize, you need to understand how AI systems decide which brand to recommend. The process differs fundamentally from classic SEO:

Dimension Google SEO (classic) AI Search (GEO)
Result 10 blue links, the user chooses 1 answer, the AI chooses for the user
Ranking factor Backlinks, keywords, technical SEO Entity recognition, source authority, content consistency
Optimization Keyword density, meta tags, page structure Clear statements, citability, fact structure
Competition Position 1–10 (everyone visible) Recommended or not (binary)

The key insight: AI systems recommend brands that exist as entities in their knowledge space — with clear attributes, consistent claims across many sources, and an unambiguous position.

GEO: Generative Engine Optimization — the 7 Strategies

1. Become an Entity, Not Just a Website

AI systems think in entities, not URLs. An entity is a clearly defined thing — a company, a product, a person — with unambiguous attributes.

Concretely, that means:

Our AI Visibility Check (German case study) shows just how much a clear entity matters: Škoda scored 98 out of 100 and was mentioned in 100% of relevant queries. Volkswagen, despite far greater brand awareness, scored only 62/100 — with a 50% mention rate.

2. Write Citable Statements

AI systems cite content that is clear, concise, and fact-based. The difference between a statement that gets cited and one that gets ignored:

The formula: concrete subject + measurable claim + source reference. This pays off measurably: expert citations increase visibility in AI answers by roughly +30%, and embedded statistics by up to +40%. Every paragraph on your site should contain at least one sentence an AI could quote directly.

3. Create an llms.txt File

Just as robots.txt tells search crawlers what they’re allowed to index, llms.txt tells AI systems what your brand is. An llms.txt file is a machine-readable short profile of your company:

llms.txt isn’t an official standard yet — but early adoption gives you an edge once AI systems start reading these files systematically.

4. Answer the Questions AI Users Actually Ask

AI users phrase questions differently than Google users. Instead of “neuromarketing agency Hamburg,” they ask: “What tools do marketers use to predict unconscious buying decisions?”

Create content that directly answers these natural-language questions — ideally in FAQ formats using <details> elements or FAQ schema. That increases the likelihood your content gets used as the source answer.

AI-Visibility Keynote & Workshop: Dr. Jonathan Mall delivers keynotes and workshops on AI visibility & GEO — half-day or full-day, in German, English, or Dutch. In the workshop format, your team works live on your brand’s GEO positioning, tests it in ChatGPT and Perplexity, and leaves with a concrete action plan. Fee range: mid-four-figure bracket (EUR). Request a keynote or workshop date →

5. Build Authority Through Third-Party Sources

AI systems weight statements more heavily when they’re corroborated by multiple independent sources. That means:

Cialdini calls this the authority principle: we trust sources that are referenced by other trustworthy sources (Cialdini, 2006, ch. 6). AI systems work on the same principle — expert citations alone increase visibility in AI answers by roughly +28%.

6. Use Digital Twins to Test AI Visibility

This is where it connects back to Digital Twins: you can ask synthetic audiences whether they’d recommend your brand — and why or why not. That gives you an early signal for how your positioning lands with AI systems, before you measure it in a real AI visibility check.

7. Invest in Answer Engine Optimization (AEO)

AEO is the SEO counterpart for AI search systems. The core principles:

Case Study: What the AI Visibility Check Found Across 293 Brands

For our AI Visibility Check (German case study), we systematically scanned 293 brands in ChatGPT and Perplexity. The average score was just 59 out of 100 points, and the brands fell into three clear groups:

The average mention rate across all 293 brands was 42%. In a direct comparison, the gap between clear and diffuse positioning stood out sharply: Škoda reached 98/100 with a 100% mention rate, while Volkswagen, despite being the larger brand, scored only 62/100 with a 50% mention rate.

The lesson: this isn’t primarily about who has the better products. What matters is who is positioned more clearly and consistently in the AI’s training data and real-time sources.

What This Means for Your Marketing Strategy

AI visibility isn’t a project you do once and check off. It’s an ongoing process that runs in three phases:

  1. Audit: Ask ChatGPT, Perplexity, and Gemini about your brand, your industry, and your competitors. Document what gets recommended and what doesn’t.
  2. Optimization: Implement the 7 GEO strategies. Prioritize entity maintenance, citable statements, and FAQ content.
  3. Monitoring: Repeat the audit every quarter. AI models are updated constantly — your visibility can shift.

The companies investing in GEO today will be the recommendations tomorrow. The ones that wait will find that their competitors are already the answer.

Further Reading

Frequently Asked Questions

What is GEO (Generative Engine Optimization)?

GEO is the practice of optimizing content, entities, and data for AI-based search systems like ChatGPT, Perplexity, and Google AI Overviews. Instead of keywords and backlinks, the focus shifts to entity recognition, citable statements, and source authority. The goal is to be named as the answer by the AI — on average, ChatGPT recommends only 2.8 brands per query.

How is GEO different from classic SEO?

Significantly. SEO optimizes for rankings on a results page with ten links that the user compares themselves. GEO optimizes for whether the AI names your brand at all as one of a handful of answers — a binary outcome: recommended or invisible. GEO doesn’t replace SEO; it adds entity maintenance, citability, and structured data on top of it.

Is there research showing that AI citations drive website visits?

Direct clicks from AI answers are rarer than classic search clicks — but the ones that do come through are more valuable. Analyses show that visitors arriving via AI recommendations convert at roughly an 11x higher rate than classic search traffic and spend about 68% more time on the page. Being cited means less traffic, but significantly higher-quality traffic.

How long does it take for GEO measures to show results?

Early effects — from new structured data or citable statements, for instance — can become visible within a few weeks, depending on the crawling and training cycles of the respective AI systems. Building a durable, consistent entity positioning, however, is an ongoing process spanning several quarters, not a one-time project.

Sources & Further Reading

  1. Aggarwal, P. et al. (2024). GEO: Generative Engine Optimization. arXiv preprint. arxiv.org/abs/2311.09735
  2. Cialdini, R. B. (2006). Influence: The Psychology of Persuasion. Harper Business.
  3. Gartner (2025). Predicts 2025: Search and AI Will Fundamentally Transform Marketing.
  4. BrightEdge (2025). State of AI in Search Report.
  5. Kahneman, D. (2011). Thinking, Fast and Slow. Farrar, Straus and Giroux.
  6. Schema.org — structured data for search engines and AI systems.
  7. llmstxt.org — the llms.txt specification.
  8. Jonathan Mall (2026). The Big AI Visibility Check — a scan of 293 brands in ChatGPT and Perplexity (German).

AI-Visibility Keynote & Workshop: Want to know how visible your brand is in ChatGPT today — and exactly what to change? Dr. Jonathan Mall delivers keynotes and workshops on AI visibility & GEO, half-day or full-day, in German, English, or Dutch. Fee range: mid-four-figure bracket (EUR). Request a date →

Dr. Jonathan T. Mall

Cognitive neuropsychologist, AI entrepreneur, and CIO of neuroflash. Jonathan analyzes how brands become visible in AI search — and published the AI Visibility Check (293 brands, VW vs. Škoda), which showed why clear positioning matters more than brand size. He has also written the in-depth AI Visibility 2026 guide for neuroflash. LinkedIn · Request a keynote or workshop