What is Answer Engine Optimisation (AEO)?

AEO is the practice of structuring your online presence so AI assistants — ChatGPT, Perplexity, Google AI Overviews — cite your business when answering relevant queries. Unlike SEO, AI engines select sources, not pages.

How does Citura calculate my AI visibility score?

Citura analyses 6 dimensions: robots.txt/llms.txt configuration, structured data (schema.org), answer-first content depth, agent.json presence, and active citation signals. Each dimension is weighted; the aggregate is your 0–100 score.

How long before I see results with Citura?

AI citation indexes update slowly — expect meaningful movement in 4–8 weeks. Your score improves within days as optimisations are deployed; actual citations follow as AI engines re-crawl your domain.

Does Citura guarantee AI citations?

No. Citura sells the system and the signals — not a placement. Anyone promising "rank #1 in ChatGPT" is misleading you. We make your business structurally citable; the AI engine decides when to cite.

Is my data safe with Citura?

Citura only analyses publicly accessible data on your domain. No credentials are required. All processing runs on EU infrastructure (Google Cloud Run, region europe-west6).

Does ChatGPT always cite sources?

Only when web search is enabled. ChatGPT without web browsing draws from its training data (knowledge cutoff August 2023 for GPT-4o). With web search enabled — available in ChatGPT Plus and the default in ChatGPT Search — it fetches live pages and cites sources inline. Optimizing for citations requires targeting ChatGPT with web search active.

How do I check if GPTBot can access my site?

Fetch your robots.txt at yourdomain.com/robots.txt and look for GPTBot. If it is not mentioned, the default depends on your server — most sites allow it by default. If you use a CMS or security plugin that adds a blanket disallow, check and override it. You can also use OpenAI's GPTBot documentation page to verify the crawler's current IP ranges.

How quickly does ChatGPT pick up changes after I optimize?

ChatGPT's web browsing index refreshes continuously but unevenly. Changes to frequently-crawled domains appear within days; less-crawled domains can take weeks. After implementing structural changes, manually query ChatGPT with your target questions weekly to track progress. Automated monitoring (Citura Pro plan) runs this systematically.

Can I pay OpenAI to be cited?

No. OpenAI has no advertising or placement product for ChatGPT citations. The system selects sources algorithmically based on relevance and quality signals. Any service claiming to "guarantee ChatGPT citations" through a fee-based submission process is misrepresenting how the system works.

How to Get Your Business Cited by ChatGPT

ChatGPT cites sources when web search is enabled. This guide explains the exact structural, technical, and content changes that increase the probability your business appears as a cited source in ChatGPT answers.

ChatGPT cites your business when it finds a page that directly answers the user's query, comes from a crawlable domain with clear authorship signals, and provides factual content with verifiable specificity. The four structural changes below have the highest observed impact on ChatGPT citation frequency.

How ChatGPT selects sources

When a ChatGPT user has web search enabled (ChatGPT Plus, GPT-4o with browsing), the system runs a search query, fetches candidate pages, and synthesizes an answer from the most relevant passages. The source URL appears as an inline citation. The selection logic prioritizes: (1) answer-first content — the page provides a complete, direct answer near the top; (2) clean HTML structure with semantic headings; (3) schema.org markup that signals the content type; (4) a domain that is not blocked in robots.txt for GPTBot.

OpenAI's GPTBot crawler (user agent: GPTBot) is what populates the web-browsing index. If GPTBot is blocked in your robots.txt — which many sites do accidentally — your pages cannot appear as ChatGPT citations regardless of their quality.

Step 1 — Unblock GPTBot in robots.txt

Check your robots.txt at yourdomain.com/robots.txt. If you see User-agent: * Disallow: / or an explicit GPTBot block, you are invisible to ChatGPT's web browsing. Add:

User-agent: GPTBot
Allow: /

User-agent: ChatGPT-User
Allow: /

Also allow PerplexityBot, Google-Extended (for Google AIO), and ClaudeBot for full AI coverage. A single robots.txt change can unlock citations across four major AI engines simultaneously.

Step 2 — Create a llms.txt file

llms.txt is an emerging standard (proposed by Jeremy Howard, fast.ai, 2024) that provides AI systems with a structured overview of your domain in plain Markdown. It is served at yourdomain.com/llms.txt. A minimal llms.txt looks like:

# Citura
> Citura is an AI visibility platform that helps SMEs get cited by ChatGPT, Perplexity, and Google AI Overviews through automated AEO/GEO optimization.

## Key pages
- [What is AEO](https://citura.app/en/resources/what-is-answer-engine-optimization-aeo): Definition and fundamentals
- [Pricing](https://citura.app/en/pricing): Plans from CHF 199/month

AI systems that support llms.txt use it to understand your site's scope before crawling individual pages. It is particularly impactful for ChatGPT's browsing mode, which uses it to prioritize crawl targets.

Step 3 — Rewrite your key pages as direct answers

Every page targeting a question-based query ("What is...", "How to...", "Best... for...") must open with the direct answer in 20–40 words. ChatGPT excerpts the most answer-like passage — if the answer is buried in paragraph four, the excerpt will be paragraph four, stripped of context. Structure each key page as:

Direct answer (first 40 words) — the complete, standalone response to the page's title question
Why / context — why this answer is correct, with verifiable specifics
How / steps — actionable detail under H2/H3 subheadings
FAQ — 3–5 related questions with concise answers (marked up with FAQPage schema)

Step 4 — Add FAQPage and Article schema

ChatGPT's browsing mode parses JSON-LD schema to understand content type and structure. FAQPage schema on a page tells the system that the Q/A pairs are authoritative answers, not body text. Article schema with datePublished, author, and publisher signals that content is regularly maintained. Both increase selection probability for informational queries.

Step 5 — Build citation mass

ChatGPT's training data (and its browsing index) biases toward domains cited frequently across the web. If your domain appears in Wikipedia, industry directories, local chamber of commerce listings, partner websites, and press mentions, it builds what researchers call "citation mass" — a proxy for authority that AI engines use in source selection. Each legitimate external reference adds to this signal.

What does not work

Several tactics marketed as "ChatGPT optimization" have no observed effect: keyword stuffing, paying for AI-specific backlink schemes, adding invisible text for AI crawlers, or submitting URLs directly to OpenAI. ChatGPT does not have a "submission" mechanism. The only reliable path is structural: make your content directly answerable, technically accessible, and factually dense.