Marketing glossary

AEO (Answer Engine Optimization) is the practice of structuring content so that answer engines, including Google AI Overviews, voice assistants, and AI chatbots, can extract a direct answer from it. It focuses on giving a clear, self-contained response to a specific question rather than earning a ranked link.

Agentic AI describes AI systems that can pursue a goal by planning steps and taking actions across tools, rather than only generating a single response. An agent can break a task into steps, use tools such as search or code, observe the results, and adjust, working toward an outcome with limited human input.

An AI agent is an AI system that pursues a goal by planning a sequence of steps and taking actions through tools, rather than only producing a single response. It can decide what to do next based on what it observes along the way, working toward an outcome with limited human input.

An AI hallucination is when a language model generates information that is false, fabricated, or unsupported, while presenting it confidently as fact. The model is not lying in any intentional sense; it is producing plausible-sounding text that is not grounded in reality.

AI Mode is Google's AI-powered search experience that answers complex questions conversationally, using generative AI to research a query across many sources and synthesize a single response. It goes beyond a list of links toward an interactive, answer-first way of searching.

AI Overviews are AI-generated answer summaries that Google displays above traditional search results for many queries. They synthesize information from multiple web sources into a direct answer, with links to the pages used as sources.

Alt text (alternative text) is a written description of an image, added in the HTML, that screen readers read aloud and search engines use to understand the image. It serves accessibility first and image SEO second.

AOV (Average Order Value) is the average amount a customer spends in a single order. It is calculated by dividing total revenue by the number of orders over a given period, and it is a core metric for understanding and improving e-commerce profitability.

Marketing attribution is the process of assigning credit for a conversion to the marketing touchpoints that contributed to it. It answers which channels, campaigns, or interactions should be credited when a customer completes a purchase or a lead.

A backlink is a link from one website to a page on another website. Search engines treat backlinks as signals of credibility and relevance, so the quantity and, more importantly, the quality of a page's backlinks influence how well it ranks.

CAC (Customer Acquisition Cost) is the total sales and marketing cost required to acquire one new customer, calculated by dividing acquisition spend over a period by the number of new customers won in that period.

A canonical tag (rel=canonical) is a piece of HTML that tells search engines which version of a page is the preferred, or canonical, one when the same or very similar content exists at multiple URLs. It consolidates duplicate pages so ranking signals point to a single URL.

Context engineering is the practice of curating and maintaining the right set of information an AI model sees at each step of a task, rather than only phrasing a single prompt. It covers the whole information environment around the model, including instructions, retrieved documents, memory, and tools.

Core Web Vitals are a set of three metrics Google uses to measure real-world page experience: Largest Contentful Paint (LCP) for loading, Interaction to Next Paint (INP) for responsiveness, and Cumulative Layout Shift (CLS) for visual stability. They are part of how Google assesses page experience.

CPC (Cost Per Click) is the amount an advertiser pays each time someone clicks their ad. It is the core pricing unit of pay-per-click advertising on platforms like Google Ads and Meta Ads, and it directly affects how far a media budget stretches.

CRO (Conversion Rate Optimization) is the practice of increasing the percentage of visitors who complete a desired action, such as buying or submitting a form, without needing more traffic. It focuses on getting more value from the visitors you already have.

A CTA (Call to Action) is a prompt that tells a visitor what to do next, such as Buy now, Get a quote, or Subscribe. It is the bridge between attention and action, turning a passive reader into a lead or customer.

CTR (Click-Through Rate) is the percentage of people who click on a link, ad, or search result after seeing it. It is calculated by dividing clicks by impressions and multiplying by 100.

Conversion rate (CVR) is the percentage of visitors or users who complete a desired action, such as a purchase, sign-up, or inquiry. It is calculated by dividing conversions by total visitors or clicks, multiplied by 100.

Deep Research is an AI capability in which an agent autonomously runs a multi-step research process, searching many sources, reading them, and synthesizing a structured, cited report, rather than answering from memory in a single reply. It trades speed for depth, taking minutes to produce a thorough answer.

Domain Authority (DA) is a score, created by the SEO company Moz, that predicts how likely a website is to rank in search results, on a scale from 1 to 100. It is a useful third-party benchmark, but it is not a metric Google uses to rank pages.

E-E-A-T stands for Experience, Expertise, Authoritativeness, and Trustworthiness. It is the framework in Google's Search Quality Rater Guidelines for evaluating whether content and its creators are credible, with Trust described as the most important member.

Email deliverability is the ability of your emails to reach recipients' inboxes rather than being blocked or filtered into spam. It is distinct from delivery: an email can be delivered to a mail server yet still land in the spam folder, which counts as a deliverability failure.

A featured snippet is a boxed answer that Google displays at the top of the search results, above the regular listings, pulled from a page that concisely answers the query. It is often called position zero because it sits above the first organic result.

A foundation model is a large AI model trained on a broad, massive dataset that can be adapted to many different downstream tasks. Rather than being built for one job, it is a general-purpose base that is then fine-tuned or prompted for specific uses. GPT, Claude, and Gemini are foundation models.

Generative AI is a category of artificial intelligence that creates new content, such as text, images, audio, video, or code, by learning patterns from large amounts of training data. Unlike systems that only classify or predict, generative models produce original output in response to a prompt.

GEO (Generative Engine Optimization) is the practice of structuring and optimizing content so that AI-powered answer engines, such as Google AI Overviews, ChatGPT, and Perplexity, select and cite it when generating responses to user queries.

Human-in-the-loop (HITL) is a design pattern where a person reviews, approves, or corrects an AI system's outputs or actions at key points, rather than letting it run fully autonomously. It keeps human judgment in control of decisions that carry risk.

Image recognition is an AI capability that identifies and classifies what appears in an image, such as objects, people, text, logos, or scenes. It lets software interpret visual content the way it interprets text, turning a picture into information a system can act on.

JavaScript is a programming language that runs in the browser to make web pages interactive and dynamic, from menus and forms to content loaded after the page first appears. For SEO, it matters because search engines must render that JavaScript to see any content it generates.

Keyword research is the process of finding and analyzing the terms people type into search engines, so you can create content that matches what your audience is actually looking for. It is the foundation of any SEO or content strategy, because it decides what you optimize for.

A knowledge graph is a structured network of entities, such as people, places, brands, and things, and the relationships between them. Search engines and AI systems use it to understand meaning and context rather than matching keywords, which is how they know that a name refers to a specific company and what that company does.

An LLM (Large Language Model) is an AI model trained on vast amounts of text to understand and generate human language. It works by predicting the most likely next piece of text given everything so far, which lets it answer questions, write, summarize, and hold a conversation.

LLMO (Large Language Model Optimization) is the practice of influencing how large language models understand, recall, and represent a brand, both in the sources they are trained on and in the content they retrieve at query time. The goal is accurate, favorable representation whenever a model mentions your category.

llms.txt is a proposed standard file, placed at the root of a website, that offers AI systems a curated, machine-readable guide to a site's most important content in Markdown. It is meant to help language models find and use the pages that matter, rather than crawling everything.

LTV (Customer Lifetime Value, also written CLV) is the total revenue or profit a business expects to earn from a customer over the entire duration of the relationship, not just their first purchase.

MCP (Model Context Protocol) is an open standard, introduced by Anthropic in November 2024, that gives AI applications a consistent way to connect to external tools and data sources. It lets a model reach live information and take actions through a shared interface instead of a custom integration for every tool.

MER (Marketing Efficiency Ratio) is a metric that measures total company revenue against total marketing spend across all channels. Unlike channel-level ROAS, it takes a blended, business-wide view of whether marketing as a whole is producing revenue efficiently.

A meta description is a short HTML summary of a page that search engines may display as the snippet beneath the title in search results. Its main job is to earn the click, and while it is not a direct ranking factor, it strongly influences whether people choose your result.

Multimodal search is search that accepts and understands more than one type of input, such as text, images, voice, and video, processed together by a single AI model rather than text alone. It lets a person search by combining a photo and a question, or speaking instead of typing.

Neural search is a search method that uses AI models to match results by meaning rather than by exact keywords. It converts text into vector embeddings, numerical representations of meaning, and returns results whose meaning is closest to the query, even when they share no words with it.

AI orchestration is the coordination of multiple AI models, tools, and agents into a single workflow so they work together toward a goal. It manages which component runs, in what order, how information passes between them, and how their outputs combine into a result.

PPC (Pay-Per-Click) is an online advertising model in which advertisers pay a fee each time their ad is clicked. Rather than earning visits organically, a business buys them, which makes traffic immediate but ongoing only as long as the budget lasts.

Programmatic SEO is the practice of generating large numbers of search-targeted pages automatically from structured data and templates, rather than writing each page by hand. Each page targets a specific long-tail query pattern, such as a location, product, or comparison.

Prompt engineering is the practice of crafting the input given to an AI model so it produces more accurate, relevant, and useful output. Because a model's response depends heavily on how a request is phrased, small changes to a prompt can meaningfully change the result.

Query fan-out is a technique used by AI search systems, notably Google's AI Mode, that breaks a single complex query into many related sub-queries, runs them at the same time, and synthesizes the results into one answer. It lets the system research a question broadly rather than matching it to a single results page.

RAG (Retrieval-Augmented Generation) is a technique that lets a large language model retrieve relevant information from an external source before generating an answer, rather than relying only on what it learned during training. It grounds the model's response in specific, up-to-date documents.

ROAS (Return on Ad Spend) is a marketing metric that measures how much revenue a business earns for every dollar it spends on advertising. It is calculated by dividing the revenue attributed to your ads by the cost of those ads.

ROI (Return on Investment) is a financial metric that measures how much profit an investment generates relative to its cost. It is calculated by dividing net profit by the total cost of the investment, usually expressed as a percentage.

Schema markup is structured data added to a web page, using the Schema.org vocabulary, that helps search engines and AI systems understand what the content is about. It labels the meaning of information on a page so machines can interpret it reliably rather than guessing from the text.

A SERP (Search Engine Results Page) is the page a search engine returns in response to a query. It contains the organic results plus a growing set of features such as ads, AI Overviews, featured snippets, and local packs that shape where attention goes.

Topical authority is the degree to which a website is seen by search engines and AI systems as a credible, comprehensive source on a subject. It is built by covering a topic in depth across many related pages, rather than through a single article or a scattering of unrelated posts.

UGC (User-Generated Content) is any content created by customers or users rather than by the brand itself, including reviews, ratings, photos, videos, comments, and social media posts. Its value comes from being seen as authentic and independent of the brand.

Voice search is the use of spoken queries, through assistants such as Siri, Google Assistant, and Alexa, to search the web or a device. Voice queries tend to be longer, more conversational, and more often phrased as full questions than typed searches.

White-label marketing is an arrangement where one company produces a service and another company rebrands and sells it as its own. It lets an agency offer services such as SEO, content, or advertising, delivered by a third-party provider, under the agency's own brand.

An XML sitemap is a file that lists the important URLs of a website to help search engines discover and crawl them. It acts as a map of the site for crawlers, and while it aids discovery, it is not a direct ranking factor in itself.

YMYL (Your Money or Your Life) is Google's term for content that could significantly affect a person's health, financial stability, safety, or well-being. Google holds YMYL pages to a higher standard of quality and trust because inaccurate information can cause real harm.

A zero-click search is a search that ends without the user clicking through to any website. The searcher's need is satisfied directly on the results page by answer boxes, AI summaries, knowledge panels, maps, or other built-in features.