SEO & GEO glossary

Alt text (alternative text) is an HTML attribute added to image tags to describe what the image shows. Search engines use alt text to understand image content, and it is critical for web accessibility. Good alt text is descriptive, concise, and includes relevant keywords naturally.

Anchor text is the visible, clickable words in a link. Search engines use anchor text to understand what the linked page is about. Varied, natural anchor text profiles are preferred over exact-match keyword anchors, which can appear manipulative.

The AI Visibility score blends how frequently your brand appears in AI-generated answers (presence), across how many prompts and engines (breadth), and how favourably (favourability), into a single 0–100 figure for the selected window. It is a composite our scorers compute over your tracked prompts — not a number any AI vendor publishes. On a connected workspace it reflects your real captures; on sample data it is an example.

ATRP (Average True Rank Position) averages your ranking position across all tracked queries, but unlike a naive average it assigns an honest penalty position to queries where you do not rank at all, rather than silently dropping them. This stops the headline number from looking artificially strong just because your worst queries were excluded.

The authority model score is our own modelled estimate (0–100) of how authoritative a domain is, used to weight which sources an AI engine is likely to trust and to size up competitors. It is a model estimate built from public signals — not a figure provided by Google or any AI vendor, and not the same as Moz DA or Ahrefs DR.

Backlinks (also called inbound links) are one of the most important ranking factors. They serve as votes of confidence from other websites. Quality matters more than quantity: a single link from a high-authority, relevant site is worth more than hundreds of low-quality links.

Bounce rate measures the percentage of single-page sessions on your site. A high bounce rate may indicate that visitors are not finding what they expected. However, a high bounce rate is not always bad - for blog posts, a user may read the entire article and leave satisfied.

A canonical URL tells search engines which version of a page is the master copy when similar or duplicate content exists at multiple URLs. This is implemented using the rel="canonical" link tag in the HTML head. It helps prevent duplicate content issues and consolidates link equity.

Content score is a composite metric that evaluates how comprehensively your content covers a topic compared to top-ranking competitors. It typically considers NLP term coverage, keyword density, readability, word count, heading structure, and semantic depth.

Core Web Vitals are three specific metrics: Largest Contentful Paint (LCP), Interaction to Next Paint (INP), and Cumulative Layout Shift (CLS). Google uses these as ranking signals. LCP measures loading speed, INP measures responsiveness, and CLS measures visual stability.

Crawl budget is the number of URLs Googlebot will crawl and index on your site within a given timeframe. For large sites, optimizing crawl budget is important to ensure your most valuable pages get discovered and indexed promptly.

CLS measures how much visible content shifts around while a page is loading. A score below 0.1 is considered good. Common causes of high CLS include images without dimensions, dynamically injected content, and web fonts causing FOIT/FOUT.

A citation is a source link or attribution an AI answer engine includes when it generates a response. Earning a citation means the engine drew on your page to answer a prompt — the AI-search equivalent of ranking. We track which prompts cite you, which cite rivals, and which sources an engine trusts so you can earn citations you are missing.

Consensus measures how much agreement there is across multiple AI answer runs (or across engines) for the same prompt — do they cite the same sources and reach the same conclusion, or does the answer swing run to run? High consensus means the answer is stable and your citation is dependable; low consensus means the result is volatile and a single capture should not be over-trusted.

Domain Authority (DA) is a search engine ranking score developed by Moz that predicts how likely a website is to rank in search engine result pages. It ranges from 1 to 100, with higher scores indicating greater ranking ability. Similar metrics include Domain Rating (DR) from Ahrefs.

Duplicate content refers to blocks of content that are either identical or very similar appearing on more than one URL. This can confuse search engines about which version to index and rank. Common solutions include canonical tags, 301 redirects, and the noindex directive.

Dwell time is the length of time between when a user clicks on a search result and when they return to the SERP. Longer dwell times generally indicate that the content was relevant and valuable. It is not an officially confirmed Google ranking factor but is widely considered influential.

Decay is when a page that used to rank or earn citations slowly slips as the content ages, competitors update, or the engine’s preferences shift. We flag decaying pages so you can refresh them before the drop becomes severe. Catching decay early is usually cheaper than recovering a page that has already fallen out of the answer set.

Defection is when an answer engine that previously cited your brand or page stops doing so and cites a competitor in your place. Tracking defection tells you where you are actively losing ground in AI answers — a more urgent signal than simply never having been cited, because it means a position you held is slipping.

E-E-A-T stands for Experience, Expertise, Authoritativeness, and Trustworthiness. These are criteria Google uses to evaluate content quality, especially for YMYL (Your Money or Your Life) topics. Demonstrating real-world experience, credentials, authoritative sources, and transparent practices improves E-E-A-T.

Featured snippets are selected search results that appear at the top of Google in a special box, often called Position Zero. They typically answer a question directly and come in formats like paragraphs, lists, tables, or videos. Optimizing for featured snippets requires clear, concise answers to common questions.

Fan-out is the process of taking a single topic and generating the realistic spread of natural-language prompts a user might ask an AI engine about it — different phrasings, intents, and follow-ups. Tracking the fan-out (rather than one canonical query) gives a truer picture of how visible you are, because AI answers vary prompt to prompt.

Find→fix is our core on-page workflow. We surface a concrete issue, let you preview the exact before/after change, apply it, and then read the page back to verify the fix actually landed. Every apply is reversible (you can undo it), which is what separates this from a one-way "auto-optimize" button.

GEO is the practice of optimizing content to be cited by AI search engines like ChatGPT, Perplexity, Google AI Overviews, and Gemini. Key techniques include BLUF formatting, data density, structured formats, definitive statements, and source attribution.

Google Search Console (GSC) is a free service that helps you monitor, maintain, and troubleshoot your site presence in Google Search results. It provides data on search queries, click-through rates, indexing status, mobile usability issues, and Core Web Vitals performance.

The go-live verdict is a per-workspace assessment of how many real data sources you have connected. Until enough are connected, surfaces stay on clearly-labelled example data rather than pretending a thin connection is a full live picture. It is the honesty gate that decides when "Example data" becomes "Live data".

Use a descriptive main heading and a logical heading structure for readers and assistive technology. A single H1 is a common editorial convention, not a Google ranking requirement; Google publishes no ideal heading count.

The hreflang attribute is used on multilingual and multi-regional websites to tell search engines which language and geographic region each page is targeting. This helps serve the correct language version to users and prevents duplicate content issues across language versions.

Half-life is an estimated time horizon over which a page is expected to lose about half of its ranking or citation value if left untouched. It is a model estimate used to prioritise refreshes — pages with a short half-life are flagged for attention sooner. It is a planning signal, not a measured guarantee.

Indexing is when search engines process and store web page information in their database so it can appear in search results. A page must be crawled and indexed before it can rank. You can check indexing status in Google Search Console and request indexing for new pages.

Internal linking is the practice of linking from one page on your website to another page on the same site. It helps search engines discover content, understand site structure, distribute page authority, and improve user navigation. A strong internal linking strategy groups related content into topic clusters.

Impressions-derived demand uses the impression counts from your connected Google Search Console as a proxy for how much real demand exists for a query, instead of relying on third-party search-volume estimates. Because it comes from your own property it reflects the demand you actually see — but it only exists once Search Console is connected.

Internal PageRank applies the PageRank algorithm to your site’s internal link graph to estimate how authority flows from page to page. Pages that many internal links point to score higher; orphan pages with no internal links score near zero. It helps you route link equity toward the pages you most want to rank.

Keyword cannibalization occurs when two or more pages on the same website target the same keyword, causing them to compete against each other in search results. This splits ranking power and can prevent either page from ranking well. Solutions include consolidating content, using canonical tags, or differentiating target keywords.

Keyword difficulty (KD) is a score, typically 0-100, that estimates how difficult it would be to rank in the top 10 for a given keyword. It considers factors like the authority and quality of currently ranking pages, number of backlinks, and competition level.

LCP measures the time it takes for the largest content element visible in the viewport to become fully rendered. A good LCP score is 2.5 seconds or less. Common causes of poor LCP include slow server response times, render-blocking resources, and large images or videos.

Link building is the practice of acquiring hyperlinks from other websites to your own. Since backlinks are a major ranking factor, earning high-quality links is essential for SEO success. Common strategies include content marketing, guest posting, broken link building, and digital PR.

Long-tail keywords are longer, more specific search phrases that typically have lower search volume but higher conversion rates. They are easier to rank for and often indicate stronger purchase or research intent. Example: "best espresso machine under $500 for beginners."

Write an accurate, page-specific summary that helps a searcher judge relevance. Google may instead select query-relevant page text, publishes no fixed character limit, and truncates snippets as needed for the device. A description can influence how a result is understood, but no click-through improvement is guaranteed.

Matched-control lift compares how pages you changed performed against a set of similar pages you left alone over the same period. It is an honest before/after comparison that controls for background trends — but it is a matched comparison, not a randomised experiment, so we label it "matched-control" and never claim it proves the change "caused" the result.

NLP is a branch of AI that helps search engines understand the meaning and context of text. Google uses NLP models like BERT and MUM to better understand search queries and page content. SEO tools use NLP to identify semantic entities and related terms that should appear in optimized content.

The noindex directive (implemented as a meta robots tag or HTTP header) tells search engines not to include a page in their index. This is useful for pages like login screens, duplicate pages, or thin content that you do not want appearing in search results.

On-page SEO refers to optimizations made directly on a web page to improve its search rankings. This includes optimizing title tags, meta descriptions, heading structure, content quality, keyword usage, internal links, image alt text, URL structure, and page speed.

Organic traffic refers to visitors who find your website through unpaid search engine results, as opposed to paid ads. Increasing organic traffic is the primary goal of SEO. It is generally considered more valuable than paid traffic because it is sustainable and indicates genuine relevance.

Page Authority is a metric that predicts the ranking strength of a specific page (as opposed to Domain Authority which evaluates the entire domain). It is influenced by the number and quality of backlinks pointing to that specific page.

Page speed measures how quickly the content on a page loads. It is a confirmed Google ranking factor and significantly impacts user experience. Key metrics include Time to First Byte (TTFB), First Contentful Paint (FCP), and Largest Contentful Paint (LCP). Common optimizations include image compression, caching, and code minification.

People Also Ask (PAA) is an interactive Google search feature that displays a list of questions related to the search query. Each question expands to show a brief answer sourced from a web page. Targeting PAA questions in your content can help capture additional SERP real estate.

Position Zero refers to the featured snippet that appears at the very top of Google search results, above the first organic result. Content that appears in Position Zero gets significantly higher click-through rates. Optimization strategies include answering questions concisely and using structured formatting.

The prompt universe is the complete set of natural-language prompts you track for a brand or topic — the fanned-out spread of phrasings and intents users actually type into AI engines. Visibility is measured across this universe rather than a single query, so the score reflects the breadth of how people really ask.

A referring domain is a unique website that contains at least one backlink to your site. Having more unique referring domains is generally more valuable than having many links from the same domain. Monitoring referring domains helps track link building progress and identify toxic backlinks.

Robots.txt is a text file placed in the root directory of a website that instructs search engine crawlers which URLs they can and cannot access. It helps manage crawl budget and prevent indexing of sensitive or duplicate pages. However, it does not guarantee pages will not be indexed.

Schema markup (or structured data) is code added to a web page that helps search engines understand its content better. It uses a standardized vocabulary from Schema.org. Common types include Article, FAQ, HowTo, Product, and Local Business. Schema can enable rich results like star ratings, FAQs, and event details in search results.

Search intent (or user intent) is the reason behind a search query. The four main types are: informational (seeking information), navigational (looking for a specific site), commercial (researching before buying), and transactional (ready to buy). Matching content to search intent is critical for ranking.

Search volume is the estimated average number of times a keyword is searched per month. Higher search volume indicates more potential traffic but usually also means more competition. It is a key metric for keyword research and content planning.

Semantic SEO is the practice of building more meaning and depth into web content. Instead of focusing solely on individual keywords, semantic SEO involves covering topics comprehensively, using related entities and terms, and structuring content to answer multiple facets of a user query.

SERP stands for Search Engine Results Page. It is the page displayed by a search engine in response to a query. Modern SERPs include organic results, paid ads, featured snippets, knowledge panels, People Also Ask, image packs, video carousels, and local packs.

A sitemap is an XML file that lists all the URLs on your website, along with metadata about each URL (when it was last updated, how often it changes, its relative importance). Submitting a sitemap to search engines helps them discover and crawl your pages more efficiently.

A slug is the portion of a URL that identifies a specific page in a human-readable form. For example, in "example.com/seo-glossary", the slug is "seo-glossary". SEO-friendly slugs are short, descriptive, include the target keyword, and use hyphens to separate words.

Share of voice (SoV) is the proportion of tracked AI answers in which your brand is mentioned or cited, compared with the rival brands you track, across the selected engines and time window. It is a relative measure: a higher SoV means you own more of the answer space than your competitors. On example/sample data it is illustrative; on a connected workspace it is computed from your real answer captures.

Stability describes how consistent a result is when the same prompts are run again — a stable citation appears reliably, an unstable one flickers in and out. We surface stability so you can tell a dependable win from a lucky single capture. It pairs with the Wilson interval: few runs plus low stability means "treat this number with caution".

Technical SEO refers to the non-content optimizations that help search engines access, crawl, interpret, and index your website. It includes site speed, mobile-friendliness, crawlability, indexing, structured data, security (HTTPS), URL structure, and XML sitemaps.

TF-IDF (Term Frequency-Inverse Document Frequency) is a statistical measure that evaluates how relevant a word is to a document relative to a collection of documents. SEO tools use TF-IDF analysis to identify terms that top-ranking pages use frequently and that your content may be missing.

Thin content refers to web pages that have very little useful, original content. This includes pages with just a few sentences, auto-generated content, or doorway pages. Google may penalize sites with significant amounts of thin content. The solution is to add depth, remove or noindex thin pages, or consolidate them.

Use a unique, concise title that accurately describes the page. Google generates title links automatically from several sources, may rewrite or truncate them to fit the device, and publishes no fixed character limit. Avoid keyword stuffing or padding a title merely to hit a length target.

Topic clusters are a content organization strategy where a comprehensive pillar page covers a broad topic, and multiple cluster pages cover specific subtopics. All cluster pages link back to the pillar page and to each other. This structure helps search engines understand topical authority and improves internal link architecture.

URL structure refers to how web addresses are formatted and organized. SEO-friendly URLs are short, descriptive, include target keywords, use hyphens as separators, avoid parameters and session IDs, and follow a logical hierarchy that mirrors your site architecture.

User experience encompasses all aspects of how users interact with your website. Good UX improves engagement metrics like time on site and pages per session, which can indirectly improve rankings. Key factors include page speed, mobile responsiveness, navigation, readability, and accessibility.

A Wilson confidence interval is a way to express how certain we are about a proportion (for example a citation rate) given how many observations it is based on. With few runs the interval is wide — meaning "we can’t be sure yet"; with many runs it narrows. We show it so a rate computed from three runs is never presented with the same confidence as one from three hundred.

An XML sitemap is a file that lists all important URLs on your website in a format that search engines can easily read. It includes metadata like last modification date, change frequency, and priority. Submitting XML sitemaps through Google Search Console helps ensure all your pages are discovered and indexed.

YMYL (Your Money or Your Life) refers to topics that could impact a person's health, financial stability, safety, or happiness. Google applies stricter quality standards to YMYL content. Pages covering medical advice, financial planning, legal information, or news require strong E-E-A-T signals to rank.