Resources
AI Visibility Glossary
Short definitions of the terms you meet in AI visibility work, from GEO to llms.txt. Where a source exists, a link sits next to the term.
AI Visibility Turkey is the Turkish-market home of Zen Media, Optimum7 and ecommerce.com.
The definitions are our working definitions, not an industry standard
For terms marked "Our working definition" there is no single industry-wide definition; we give the meaning we use. Definitions that rest on an engine’s or organization’s documentation carry a source link. For wider context see the AI visibility guide.
- AEO (answer engine optimization)
- Building content and page structure that gives a direct, short, quotable answer to a question. (Our working definition.)
- Agentic Commerce Protocol
- A protocol open-sourced by OpenAI with Stripe, aimed at checkout inside ChatGPT (as reported). Launch coverage was US-first; availability in Turkey is unverified. TechCrunch ↗
- AI Mode (AI Modu)
- Google’s AI search experience, which handles queries with query fan-out and can show a wider set of links than classic search. Google Search Central ↗
- AI Overviews (AI Bakışı)
- The AI summary shown above some Google searches; it displays supporting links. Per Google it appears only when its systems judge it useful, so it often does not trigger. Google Search Central ↗
- In how many tracked prompts your brand is mentioned compared with named competitors. No engine publishes this metric in full. (Our working definition.)
- Bing AI Performance
- A report in Bing Webmaster Tools (public preview) showing citations across Microsoft Copilot, Bing AI summaries and partner integrations: total citations, average cited pages, grounding queries (sample), page-level citations and visibility trends. It gives citation counts only, not placement. Bing Webmaster Blog ↗
- ChatGPT-User
- The OpenAI bot that fetches pages on a user’s request. robots.txt may not apply to it. OpenAI: bots ↗
- Citation (atıf)
- The AI naming your company in its answer; engines often also show a source link. (Our working definition.)
- Claude-SearchBot
- The Anthropic bot that indexes for search relevance. Per Anthropic, blocking it may reduce visibility in Claude search answers. Anthropic: web crawling ↗
- Claude-User
- The Anthropic bot that fetches pages when a user’s question needs it. Blocking it may reduce visibility in Claude search answers. Anthropic: web crawling ↗
- ClaudeBot
- The bot Anthropic uses for training data. Anthropic: web crawling ↗
- Entity (varlık)
- A brand, person or concept recognized by search and AI systems as one consistent thing. Consistent names and profile details across platforms support this. (Our working definition.)
- GEO (generative engine optimization)
- The work of preparing content so that generative AI answers use it as a source. The term was formalized in a 2023 paper by Aggarwal and colleagues. (Our working definition.) arXiv 2311.09735 ↗
- Google-Extended
- A robots.txt token that controls Gemini training and grounding. It does not affect inclusion in Google Search. Google: common crawlers ↗
- GPTBot
- A bot OpenAI may use for training. It is independent of the search setting. OpenAI: bots ↗
- Grounding
- A model basing its answer on real sources it has retrieved. The Google-Extended token covers Gemini training and grounding. (Our working definition.)
- Hallucination
- An AI stating something untrue about a brand. For reputation, it is a kind of error that needs correcting. (Our working definition.)
- hreflang
- Markup that tells a search engine about the language or regional versions of the same page. It is used to pair the Turkish and English versions. (Our working definition.) Google: localized versions ↗
- Knowledge graph (bilgi grafı)
- A data network that stores entities and the relationships between them in structured form. Wikidata is one open example. (Our working definition.)
- LLM (large language model)
- An AI model trained on large amounts of text that produces answers as text. Assistants such as ChatGPT, Gemini, Claude and Grok are built on models of this kind. (Our working definition.)
- LLM SEO
- Another name for the same idea as GEO and AEO: the work of being visible in assistants built on large language models. (Our working definition.)
- llms.txt
- An informal proposal by Jeremy Howard, first published on 3 September 2024; it is not a ratified standard. As reported, Google’s John Mueller has said no AI system uses it. We treat it as optional hygiene, not a ranking lever. llmstxt.org ↗
- Mention (bahsedilme)
- Your brand name appearing in the text of the answer; it does not have to be shown as a source. (Our working definition.)
- Merchant Center
- Google’s product data platform; per Google it is the core data source for AI Mode and Gemini shopping. The Buy button also needs the native_commerce attribute and an interest form. Google Merchant Center Help ↗
- Non-branded citation
- The AI choosing you even though your company name was not in the buyer’s question. A branded question contains your name and does not count here. (Our working definition.)
- OAI-SearchBot
- The bot OpenAI uses to surface sites in ChatGPT search. Sites that do not allow it do not appear in ChatGPT search answers. Setting changes take about 24 hours to take effect. OpenAI: bots ↗
- Perplexity-User
- The Perplexity bot that fetches pages on a user’s request; it generally ignores robots.txt. Perplexity: bots ↗
- PerplexityBot
- The bot that surfaces and links sites in Perplexity search; Perplexity says it is not used to crawl for foundation-model training. Perplexity: bots ↗
- Prompt (istem)
- A question or command typed into an AI assistant. In measurement, prompts are built from the questions buyers really ask. (Our working definition.)
- Prompt universe
- The full set of possible buyer questions for a category, with persona and intent segments. (Our working definition.)
- Query fan-out (sorgu yayılımı)
- AI Mode splitting a question into subtopics and running many related searches together. This is how Google describes the behavior. Google Search Central ↗
- robots.txt
- A file in your site’s root that tells bots which pages they may crawl. For AI bots, search and training permissions can be set separately. (Our working definition.)
- schema.org
- The shared vocabulary for structured data; it defines types such as Organization, Product and Article. Google says no special schema is needed for AI features. (Our working definition.) schema.org ↗
- Structured data (JSON-LD)
- Markup that labels page content in a machine-readable way; Google recommends JSON-LD. The markup must match visible content; fake or irrelevant markup risks a manual action. Valid markup does not guarantee display. Google: structured data policies ↗
- Wikidata
- Wikimedia’s open, structured database. An item needs a sitelink to a Wikimedia project, a clearly identifiable entity with serious public references, or a structural need. Wikipedia requires significant coverage in reliable independent sources; we do not sell "getting a Wikipedia page". Wikidata: Notability ↗
Where does your brand stand in AI answers?
Send us your website and topics and we will reply by email.