• 301 RedirectA 301 redirect is an HTTP status code that tells browsers and search engines a page has permanently moved to a new URL. Google treats it as a signal to make the new URL canonical, replacing the old address in search results and passing the old URL's accumulated link equity (ranking) to the destination.SEO
• AEOAEO (Answer Engine Optimization) is the practice of structuring and strengthening your content so that answer engines like ChatGPT, Perplexity, and Google's AI Mode cite, summarize, or mention it when they generate a direct answer to a user's question. Unlike SEO, which aims for a high spot in a ranked list of links, AEO aims to get your content included in the answer itself even when nobody clicks through.GEO & AI Search
• Agentic RAGAgentic RAG is a form of retrieval-augmented generation in which an autonomous AI agent decides whether and how to retrieve, rewrites queries, calls tools, evaluates the results, and retrieves again when needed. Unlike standard RAG, which retrieves once and then generates an answer, it turns retrieval into a dynamic process of iteration, planning, and verification.GEO & AI Search
• Agentic SearchAgentic search is a search paradigm in which an AI agent breaks a single question into smaller sub-tasks and then plans, searches, and verifies results in a loop to construct an answer on its own. Unlike traditional search, which returns a list of documents matching a query in one pass, it cross-checks multiple sources and reasons about what to look for next, assembling the answer step by step.GEO & AI Search
• Agentic WebThe Agentic Web is the emerging web paradigm in which AI agents, acting on a user's delegated intent, autonomously browse, coordinate, and even transact across the web instead of people clicking through pages themselves. It marks the shift in the web's center of gravity from human-driven interaction to machine-to-machine interaction between agents.GEO & AI Search
• AI AgentAn AI agent is an autonomous system in which a large language model (LLM) plans its own steps toward a goal, calls tools, and uses feedback from its environment to iteratively decide what to do next. Unlike a workflow that follows a predefined code path, an agent hands the LLM direct control over how the task gets done.GEO & AI Search
• AI CitationAn AI citation is the source link or footnote that a generative search engine — such as ChatGPT, Perplexity, or Google's AI Overviews — displays to credit the web pages it drew on when generating an answer. The goal is no longer to rank as a blue link, but to be selected as a cited source inside the AI-generated answer itself, which makes citations the ultimate success metric that GEO and AEO aim for.GEO & AI Search
• AI Content GenerationAI content generation is the practice of using artificial intelligence tools such as large language models (LLMs) to produce content like text and images on an automated or semi-automated basis. From an SEO standpoint, Google evaluates content by how helpful and high-quality it is for users rather than by whether it was made with AI, so what matters is the quality and usefulness of the output, not the method of production.GEO & AI Search
• AI CrawlerAn AI crawler is an automated bot that collects web pages to train large language models (LLMs) or to generate AI search answers. Prominent examples include OpenAI's GPTBot, Anthropic's ClaudeBot, and Google-Extended, each identifiable and controllable through its own User-Agent and robots.txt token.GEO & AI Search
• AI CrawlingAI crawling refers to the act and overall process by which AI systems such as ChatGPT, Gemini, and Perplexity automatically gather and read web pages to train models, build search indexes, and answer user questions in real time. Unlike traditional search crawling, which indexes pages to rank them, the defining difference is that the collected content becomes the raw material for training data or generated answers.GEO & AI Search
• AI OverviewsAI Overviews is a Google Search feature that places a generative-AI summary at the very top of the results page alongside links to the web pages it drew from. A custom Gemini model synthesizes multiple sources to answer the core of a query; it launched broadly in the U.S. in May 2024 and has since expanded worldwide.GEO & AI Search
• AI SearchAI Search is a search experience in which generative AI reads and synthesizes web documents to write a direct answer to your question, instead of returning a list of links. Google AI Overviews, ChatGPT search, Perplexity, and Gemini are leading examples, and most run on a RAG (retrieval-augmented generation) architecture.GEO & AI Search
• AI SlopAI slop is a derogatory term for low-quality content churned out quickly and at scale by generative AI, with little regard for accuracy or value. It spans text, images, video, and audio, and typically exists to chase clicks and revenue in the attention economy.GEO & AI Search
• AI Trust SignalsAI trust signals are the verifiable cues that generative search systems like ChatGPT, Perplexity, and Google's AI Overviews use to decide which sources are reliable enough to cite in their answers. Common signals include clear authorship, supporting evidence such as citations and statistics, consistent entity information across the web, structured data, and third-party mentions.GEO & AI Search
• AI VisibilityAI Visibility is a metric that captures how often, and in what context, your brand, products, or content get mentioned or cited when generative AI tools such as ChatGPT, Perplexity, and Google AI Overviews compose their answers. Unlike traditional SEO, which measures ranking on a results page, it measures how present you are inside the AI-generated answer itself.GEO & AI Search
• Answer EngineAn answer engine is a system that, instead of returning a list of links, synthesizes information from multiple sources to generate a complete, direct answer to a question. ChatGPT, Perplexity, Google AI Overviews, and voice assistants are leading examples, letting users get instant answers without clicking through.GEO & AI Search
• BacklinkA backlink is a hyperlink from a page on another website that points to your page. Search engines like Google read backlinks as votes of confidence, using them to gauge a page's authority and relevance and to discover new pages, which makes them a core off-page SEO signal.SEO
• Brand MentionA brand mention is any reference to a brand name, product name, or related term in web content, regardless of whether a link is attached. Mentions are either linked (with a hyperlink) or unlinked (text only), and in the AI search era even link-free references contribute to brand visibility.GEO & AI Search
• Canonical URLA canonical URL is the version a search engine selects as the representative (master) page among several URLs that serve the same or near-identical content. Site owners can signal a preference with a rel="canonical" tag, an HTTP header, a sitemap, or a 301 redirect, but Google makes the final choice by weighing multiple signals.SEO
• Chain of ThoughtChain of Thought (CoT) is a prompting technique that boosts complex reasoning in large language models by getting them to spell out intermediate reasoning steps before committing to a final answer. Instead of jumping straight to the result, the model works through the problem, which sharply improves accuracy on arithmetic, commonsense, and symbolic tasks.GEO & AI Search
• ChunkingChunking is the process of splitting long documents into smaller units (chunks) suited to retrieval and embedding in a RAG pipeline. Because the split method, chunk size, and overlap directly shape retrieval accuracy and answer quality, it is treated as a core step in RAG pipeline design.GEO & AI Search
• Citation OptimizationCitation optimization is the practice of designing content so that generative search engines like ChatGPT, Perplexity, and Google's AI Overviews cite it as a source when composing their answers. Unlike traditional SEO, the goal is not a search ranking but being cited and linked inside the AI-generated answer itself.GEO & AI Search
• Context EngineeringContext engineering is the discipline of designing and managing the context an LLM receives (instructions, retrieval results, tools, memory, and conversation history) to elicit the best possible output. It is a broader concept that subsumes prompt engineering's focus on wording, addressing what to place in context, in what format, and when.GEO & AI Search
• Context WindowA context window is the maximum span of input and output tokens a large language model (LLM) can reference together in a single request. It functions as the model's working memory, and anything beyond this limit gets truncated or goes unprocessed.GEO & AI Search
• Conversational SearchConversational search is a search paradigm where you ask in natural-language sentences instead of stringing together keywords, then exchange follow-up questions that carry context forward to satisfy an evolving information need. The defining trait — seen in Google's AI Mode and tools like ChatGPT and Perplexity — is that an answer is narrowed across multiple conversational turns rather than a single query.GEO & AI Search
• EmbeddingAn embedding is a representation that converts data such as words, sentences, or images into a real-valued vector of hundreds to thousands of dimensions while preserving meaning. The closer two items are in meaning, the closer their vectors sit, which enables search and comparison based on meaning rather than keywords and underpins semantic search and RAG.GEO & AI Search
• Entity-Based SEOEntity-Based SEO optimizes for the real-world things behind a search query — people, places, objects, and concepts — and the relationships between them, rather than for the query string itself. The goal is to have search engines recognize your brand, people, and products as distinct entities within Google's Knowledge Graph.SEO
• Featured SnippetA featured snippet is a Google search feature that displays an extracted answer to a user's query in a box at the very top of the results, often called position zero. Unlike a standard result, it surfaces the descriptive snippet first and flips the usual page format.SEO
• Fine-TuningFine-tuning is the practice of taking a pre-trained model that has already learned from large amounts of data and training it further on task- or domain-specific data to adjust its weights. It is used to bake a desired tone, format, and expertise directly into the model.GEO & AI Search
• Function CallingFunction calling is a capability that lets an LLM read the schema of a developer-defined function or tool and produce the arguments needed to call it as structured data in a fixed JSON format. The model does not execute the function itself; it is the application's job to take the generated arguments and run the actual code.GEO & AI Search
• GEOGEO (Generative Engine Optimization) is the practice of optimizing your content so that generative engines like ChatGPT, Perplexity, and Google's AI Overviews cite and surface it when they synthesize an answer. Unlike SEO, which targets rankings in a list of links, GEO aims to get your brand included inside the AI-generated answer itself.GEO & AI Search
• Google AI ModeGoogle AI Mode is Google's generative search experience: instead of a list of links, you get a unified answer synthesized by Gemini, along with follow-up questions and supporting references. It handles complex queries through "query fan-out," splitting a single question into multiple sub-questions that are searched simultaneously.GEO & AI Search
• Grounded GenerationGrounded generation is a generation method in which a model produces answers based on retrieved or supplied evidence rather than relying solely on its own parametric memory, while citing the sources behind each claim. It corresponds to the 'generation' step of a RAG pipeline and aims to reduce hallucination and improve verifiability by tying each statement back to its supporting documents.GEO & AI Search
• GroundingGrounding is the concept and process of tying an LLM's output to verifiable external sources and facts, reducing hallucination and making answers checkable. It focuses on the act of anchoring a response to factual evidence — such as search results or trusted documents — rather than relying on training data alone.GEO & AI Search
• HallucinationA hallucination is when a large language model (LLM) generates content that is false or unsupported by evidence, yet presents it as if it were accurate. Because the output is grammatically fluent and stated with confidence, the errors are easy to miss.GEO & AI Search
• Hybrid SearchHybrid search runs a keyword search (a sparse-vector method like BM25) and a semantic vector (embedding) search in parallel, then fuses the two result sets into a single ranking. Because it captures exact term matches and contextual meaning at the same time, it is widely used to improve retrieval accuracy in RAG and AI search.GEO & AI Search
• JSON-LDJSON-LD is a W3C standard format for expressing Linked Data in JSON, and it is one of the ways to embed structured data into a web page. Google recommends JSON-LD over the other two structured data formats, Microdata and RDFa.SEO
• Knowledge CutoffA knowledge cutoff is the date marking the end of the data a large language model (LLM) was trained on; events and information created after that point are not part of the model's own knowledge. Anything more recent has to be supplied through external tools like web search or RAG, or the model simply won't know it accurately.GEO & AI Search
• Knowledge GraphA knowledge graph is a knowledge base that represents entities — people, places, and things — as nodes and the relationships between them as edges, enabling Google to interpret queries as concepts rather than literal keyword strings. Google introduced its own version in 2012 under the slogan "Things, not strings."GEO & AI Search
• Knowledge PanelA Knowledge Panel is the box that appears on the right side (desktop) or at the top (mobile) of Google's search results when you search for an entity such as a person, place, organization, or thing, summarizing that entity's key information. It is generated automatically from Google's Knowledge Graph and surfaces details like the name, description, image, key facts, and social profiles at a glance.GEO & AI Search
• LLMA large language model (LLM) is an AI model pre-trained on vast amounts of text that understands and generates natural language by predicting the next word (token) probabilistically. LLMs are the engines behind services like ChatGPT, Gemini, and Claude, and behind Google's AI Overviews, sitting at the center of search's shift from a list of links to AI-synthesized answers.GEO & AI Search
• LLM VisibilityLLM Visibility measures how often and how prominently a brand surfaces inside the answers generated by large language models such as ChatGPT, Claude, Gemini, and Perplexity. It tracks direct mention, citation, and recommendation within the AI's answer text rather than the ranking of blue links in search results.GEO & AI Search
• LLMOLLMO (Large Language Model Optimization) is a marketing practice that optimizes content and brand signals so a company is mentioned, cited, and recommended more often inside answers generated by LLM-based tools such as ChatGPT, Gemini, Claude, and Perplexity. Its goal is exposure and citation within AI-generated conversational answers rather than a higher search ranking.GEO & AI Search
• llms.txtllms.txt is a proposed standard: a Markdown file placed at a website's root (/llms.txt) that gives large language models a curated summary and links to the site's most important documents so they can understand it at inference time. Proposed by Jeremy Howard of Answer.AI on September 3, 2024, it differs from robots.txt — instead of controlling crawler access, its goal is to gather AI-friendly content in one place and point models to it.GEO & AI Search
• Mixture of ExpertsMixture of Experts (MoE) is a neural network architecture that uses several specialized expert sub-networks and a router that activates only a few of them for each input token. It lets a model grow its total parameter count dramatically while keeping the actual computation per token limited to a small subset of experts.GEO & AI Search
• Model Context ProtocolThe Model Context Protocol (MCP) is an open standard, released by Anthropic in November 2024, for connecting LLM-based AI applications to external tools and data sources. Instead of building a separate connector for every data source, it links models and external systems through a single standardized interface, often described as "USB-C for AI."GEO & AI Search
• Multimodal SearchMultimodal search is a way of searching that combines inputs in several formats — text, images, voice, and video — into a single query. Unlike traditional search that relies on keyword matching, it compares inputs of different formats by meaning to find an answer.GEO & AI Search
• Prompt EngineeringPrompt engineering is the practice of designing the instructions, examples, format, and role within an input prompt so that a large language model (LLM) reliably produces the output you want. The goal is to improve response quality and consistency through the input text alone, without changing the model's weights.GEO & AI Search
• Prompt InjectionPrompt injection is a security attack in which an adversary feeds maliciously crafted input to an LLM to override or hijack the system's original instructions. It exploits the structural weakness that LLMs cannot tell trusted instructions apart from untrusted data, and it is a fundamentally different concept from prompt engineering.GEO & AI Search
• Query DecompositionQuery decomposition is a technique that breaks a single complex query into several simpler sub-questions that can each be answered independently. Because each sub-question is retrieved and reasoned over separately and the results are then combined into a final answer, it improves retrieval accuracy and answer quality on multi-hop questions whose evidence is scattered across multiple documents.GEO & AI Search
• Query Fan-OutQuery fan-out is an information retrieval technique in which generative search systems like Google's AI Mode and AI Overviews expand a single user query into several related sub-queries, run them simultaneously (in parallel), and then synthesize the results into one answer. Unlike traditional search, which matches one keyword to a single set of results, one question triggers many searches that pull in a broader, more varied range of sources.GEO & AI Search
• Query RewritingQuery rewriting is the practice of transforming a user's original query into a form that search systems can match more effectively, using techniques such as synonym expansion, typo correction, removal of unnecessary phrasing, and intent clarification before matching and scoring take place. It focuses on turning one query into a single, better query, and improves precision and recall in both search engines and RAG pipelines.GEO & AI Search
• RAGRAG (Retrieval-Augmented Generation) is a technique in which an LLM first retrieves relevant documents from an external knowledge base and then grounds its answer in that retrieved content. Instead of relying solely on the knowledge baked into its parameters, the model pulls in current or specialized material at answer time, improving accuracy and citing sources without any retraining.GEO & AI Search
• RerankerA reranker is a second-stage refinement step that re-scores the candidate documents returned by first-stage retrieval against the query and reorders them by relevance. It typically relies on a cross-encoder, which takes the query and document together as a single input to produce a precise relevance score.GEO & AI Search
• Rich SnippetA rich snippet (which Google officially calls a "rich result") is an enhanced search listing that adds extra details — such as star ratings, images, prices, or FAQs — to a normal organic result, based on structured data embedded in the page. Unlike a featured snippet, which surfaces an answer box above the results, a rich snippet enriches the organic listing itself.SEO
• RLHFRLHF (Reinforcement Learning from Human Feedback) is a technique that trains a reward model on human preference data and then optimizes that reward signal with reinforcement learning to align a large language model (LLM) with human intent and values. Its goal is to steer model behavior toward the outputs people actually prefer, even among answers that would otherwise look equally valid.GEO & AI Search
• Schema MarkupSchema markup is standardized structured data that uses the schema.org vocabulary to spell out the meaning of a page's content (author, rating, price, event, and so on) so search engines can understand it. Google reads this markup to display rich results such as star ratings, FAQs, and breadcrumbs, and although it supports JSON-LD, Microdata, and RDFa, it recommends JSON-LD because it is the easiest format to implement and maintain.SEO
• Search IntentSearch intent is the real goal a user is trying to accomplish when they type a query — the reason behind the search. It typically falls into four buckets (informational, navigational, commercial, and transactional), and both search engines and AI answer engines prioritize the content that best matches that intent, making it the most important starting point in SEO.Content & Strategy
• Semantic ChunkingSemantic chunking splits a document along meaning boundaries rather than fixed units such as character count. It is a RAG chunking strategy that locates points where the embedding similarity between adjacent sentences drops sharply and breaks there, yielding chunks that are semantically cohesive.GEO & AI Search
• Semantic SearchSemantic search is a search approach that finds relevant results by understanding the meaning, intent, and context of a query rather than matching words literally. Its core technique converts text into vectors (embeddings) and retrieves semantically similar documents, underpinning modern search from Google Hummingbird and BERT to RAG and AI search.GEO & AI Search
• Semantic SEOSemantic SEO is a content strategy that optimizes for a topic's full meaning, context, and search intent rather than repeating a single keyword, so search engines understand the content deeply and recognize topical authority. It grew in importance as Google's NLP technologies like BERT and MUM, together with the Knowledge Graph, began interpreting concepts instead of words.SEO
• Share of ModelShare of Model (SoM) measures how often, and how favorably, generative AI systems like ChatGPT and Gemini mention or recommend your brand versus competitors when answering category-related questions. It has been proposed as a new brand-visibility KPI that succeeds traditional Share of Voice now that AI has become both the source of and the recommender behind the answer.GEO & AI Search
• Structured DataStructured data is markup that describes a page's information using the schema.org vocabulary, so search engines can interpret the content precisely. This standardized data is what makes rich results — star ratings, prices, recipe details, and the like — eligible to appear in search results.SEO
• Structured OutputStructured output is a capability that forces an LLM to produce responses conforming exactly to a predefined format, such as a JSON Schema, instead of free-form text. Because the schema is enforced during token generation rather than merely requested in the prompt, the result is always parseable data with no missing required fields or malformed structure.GEO & AI Search
• SXOSXO (Search Experience Optimization) combines SEO with user experience to optimize not just search visibility but also the post-click journey of navigation, satisfaction, and conversion. Unlike SEO, which targets rankings themselves, SXO aims to make the entire user journey from search to conversion seamless.GEO & AI Search
• TokenizationTokenization is the process of breaking text down into tokens, the smallest units a language model actually processes. In English, one token corresponds to roughly 4 characters (about 3/4 of a word), and subword algorithms such as BPE split words into smaller pieces for the model to handle.GEO & AI Search
• Tool UseTool use is an LLM's ability to call external tools such as search, calculators, code execution, and APIs to extend itself beyond its own limits. It is the core behavior behind AI agents, and in practice it is implemented through the function calling mechanism.GEO & AI Search
• Topical AuthorityTopical authority is the state in which search engines and AI search treat a site as a trustworthy, expert source across an entire subject area. It is earned by covering a topic comprehensively and consistently — through topic clusters, pillar pages, and internal links — rather than by ranking for individual keywords.Content & Strategy
• TransformerA Transformer is a neural network architecture that uses a self-attention mechanism to compute the relationships among all tokens in an input sequence in parallel. Introduced in Google's 2017 paper "Attention Is All You Need," it serves as the foundation for modern large language models (LLMs) such as GPT, Claude, and BERT.GEO & AI Search
• Vector DatabaseA vector database stores high-dimensional embeddings of data such as text and images and uses approximate nearest neighbor (ANN) search to quickly find the vectors whose meaning is closest to a query vector. Unlike a traditional database that retrieves exact matches, it searches by similarity (distance).GEO & AI Search
• Vertical SearchVertical search is a specialized type of search that returns results within a single industry, content type, or domain rather than across the entire web. Platforms like Amazon for products, YouTube for video, and Zillow for real estate each focus on one field to deliver more precise, more relevant results.GEO & AI Search
• Voice SearchVoice search is a way of finding information by speaking a query out loud instead of typing, where a voice assistant such as Siri, Google Assistant, or Alexa interprets the natural-language question and typically reads or shows a single answer. Because queries tend to be longer and conversational while the result narrows to one answer, voice search is optimized differently from text search.GEO & AI Search
• Zero-Click SearchA zero-click search is one where the user finishes searching without clicking any link on the results page. Google answers directly on the SERP through features like featured snippets, knowledge panels, and AI Overviews — or the user abandons or refines the query — so no traffic flows to an external website.GEO & AI Search