How the AI Trust Audit Works
The AI Trust Audit analyzes your website the way AI systems like ChatGPT, Perplexity, or Google AI Overviews do — from the outside, without JavaScript, based on raw HTML alone. Over 70 automated checks across 7 categories evaluate whether your content is discoverable, understandable, and trustworthy for AI crawlers. Free, in seconds, no login required.
Why AI Visibility Matters Now
AI assistants increasingly recommend products and shops — but most merchants have no idea how they are perceived. If you're invisible to AI, you won't be recommended — and you'll lose customers without even knowing it.
60% use AI for shopping
Over 60% of consumers have already used AI assistants for purchase decisions. (Bloomreach, 2025)
87% of shops are invisible
87% of online stores don't appear in AI recommendations from ChatGPT & Co. at all. (StoreRank.ai, 2,000 stores)
13x more AI traffic
Generative AI traffic to e-commerce sites grew 13x in 2025. (Adobe Analytics)
From Keyword to Context
AI prompts are longer and more context-rich than traditional search queries — this changes which content gets found. Traditional keyword optimization alone is no longer enough to appear in AI recommendations.
Short Keywords
Google searches are 3–5 words. Shops optimize for individual terms.
Context-Rich Prompts
ChatGPT prompts contain 42 words with context, preferences and constraints.
Extended Sessions
AI sessions span 348 words — an entire conversation instead of a single search.
Source: Metehan AI
From search to purchase: Agentic Commerce
AI agents will soon shop autonomously for your customers. ChatGPT Operator, Perplexity Shopping and Google AI Agents research, compare and order — without a human ever visiting your website. Your shop must be machine-readable and trustworthy to be considered by these agents.
Agent as buyer
AI agents research, compare prices and reviews, and purchase autonomously — on behalf of the user, without any human interaction with your shop.
Trust decides
Agents prefer shops with clear trust signals: SSL certificates, structured data, verified reviews and transparent terms of service.
Audit as preparation
Our AI Trust Audit measures exactly the signals that AI agents evaluate — and shows you where your shop needs to improve.
How AI Bots See Websites
AI crawlers experience the web very differently from humans — and most shops aren't ready for it. What a bot can't read simply doesn't exist for AI.
Raw HTML Only
— LLM crawlers fetch initial HTML, no JavaScript execution.
No Cookies or Sessions
— Bots can't access personalized or gated content.
SPAs Are Invisible
— JS-rendered apps may appear empty to crawlers.
What Makes Websites Machine-Readable
AI systems can only use what they understand. These three building blocks make your content accessible and interpretable for crawlers.
Structured Data (Schema.org)
JSON-LD markup like Product, FAQ, Organization, or BreadcrumbList tells AI systems exactly what's on a page — price, availability, reviews. Without schema, crawlers have to guess.
Clear HTML Structure
Semantic headings (H1–H6), lists, tables, and <article> elements. AI crawlers scan page structure to distinguish main content from navigation.
Machine-Friendly Content
Explicit meta descriptions, canonical tags, clean robots.txt and sitemaps. Plus alt text for images and publication dates — everything that provides context without a crawler having to interpret.
The Multi-Level Model of AI Search
AI visibility builds on three levels. The top — semantic relevance — depends on content and context and can't be measured statically. But without the technical foundation, you'll never get there. The AI Trust Audit measures exactly the levels you can influence yourself.
Semantic Fit
Are you relevant for the prompt?
Citation Probability
Will an AI trust you?
Technical Discoverability
Can an AI read you?
AI Trust Audit Zone
AI Trust Audit Zone
The AI Trust Audit measures the bottom two levels
Technical Foundation
Can AI read your site? Crawlability, HTML structure and structured data form the foundation.
Trustworthiness
Will AI cite you as a source? Reviews, authority and external signals are decisive.
Semantic Relevance
Does your content match the user query? This level depends on context and can't be measured statically.
How It Works
Three layers of analysis working together to build a comprehensive trust profile. No access needed — we analyze from the outside, exactly as an AI bot would.
Outside-In Scraping
We see your shop exactly as a bot would — no login, no cookies, pure HTTP requests.
Regex + AI Analysis
Pattern matching for structured data (JSON-LD, meta tags, legal pages). AI for semantic understanding of content quality.
External APIs
Google Safe Browsing, WHOIS lookups, review platforms, Reddit mentions, Trusted Shops — cross-referencing multiple trust sources.
What AI crawlers actually see
A 2026 study by Writesonic tested 6 LLM crawlers and found: ChatGPT, Claude and Gemini read raw HTML — no JavaScript at all. Copilot, DeepSeek and Grok do execute JS but only wait between 500 milliseconds and 3 seconds. None of them scroll. In direct conversation, all 6 LLMs ignore the entire <head> section except the <title> tag — even JSON-LD. Our checks reflect this reality and put more weight on core content in initial HTML, a meaningful <title>, microdata and a noscript fallback.
What You Get
An automated trust score showing how AI systems would evaluate an online shop. In under 30 seconds you'll know where your shop stands — and what you can improve.
Comprehensive Analysis
70+ automated checks across 7 categories cover all relevant aspects of AI visibility.
Multilingual
Available in 8 languages — results and recommendations in your language.
Actionable Results
Results in under 30 seconds with concrete recommendations per check.