SEO Glossary

A

• AI Hallucination

  An AI hallucination is a confident, plausible-sounding output from a language model that is factually wrong, fabricated, or unsupported by the retrieved sources. Common causes: thin retrieval, ambiguous prompts, and the model's tendency to fill gaps with its training-data priors.

• AI Mode

  AI Mode is Google's standalone AI-only search interface, announced at Google I/O on May 20, 2025. Unlike AI Overviews — which sit above the classic SERP — AI Mode replaces the entire results page with a Gemini- generated conversational answer. Opt-in via a tab next to All, Images, News, etc.

• AI Overview

  AI Overview is the Gemini-generated answer block Google renders at the top of selected SERPs. Citations link to source pages. It graduated from the Search Generative Experience (SGE) preview to general availability in the US on May 14, 2024 (Google I/O), then expanded to 100+ countries through late 2024.

• AI training opt-out

  AI training opt-out is the bundle of mechanisms that prevent your content from being used to train language and image models: robots.txt blocks for named user-agents, meta `noai`/`noimageai` tags, `X-Robots-Tag` HTTP headers, and — in the EU — text-and-data-mining reservation under DSM Directive Article 4(3).

• Alt text

  Alt text is the value of the `alt` attribute on an `<img>` tag. It serves two purposes: an accessibility requirement (screen readers announce it, WCAG 1.1.1 mandates it) and an SEO signal (Google treats it as anchor text for linked images and as a relevance cue in Google Images and AI Overviews).

• Anchor text

  Anchor text is the visible, clickable text of an HTML link (`<a href="...">anchor text</a>`). It is a primary relevance signal for the target URL. Types: exact match, partial match, branded, naked URL, generic ("click here"). Over-optimization on exact-match anchors triggers spam filters.

• Anchor text diversity

  Anchor text diversity is the ratio of anchor types — branded, naked URL, exact match, partial match, generic — across a domain's external backlinks. A healthy profile is heavily branded and naked, modestly partial, lightly exact. Excessive exact-match concentration is a Penguin-era spam signal.

• Answer Engine Optimization (AEO)

  Answer Engine Optimization (AEO) is the practice of optimizing content to win direct-answer surfaces: featured snippets, People Also Ask, voice assistants, and FAQ rich results. The term predates GEO (~2017-2020) and targets extractive answers — the engine quotes a passage of yours verbatim.

• Apple Business Connect

  Apple Business Connect is Apple's free business-listing platform, launched in January 2023 to replace Apple Maps Connect. A verified listing controls how a business appears in Apple Maps, Spotlight Search, Siri suggestions, and Look Around. It's the Apple equivalent of Google Business Profile.

• Attribution model

  An attribution model is a rule for distributing conversion credit across the touchpoints in a user's journey. Common models: last-click, first-click, linear, time-decay, position-based, and data-driven. GA4 made data-driven the default in May 2023, retiring last non-direct click for new properties.

• Average position

  Average position in Google Search Console is the mean rank across all queries your URL appeared for, weighted by impressions. The lower the number the better. It is one of the most frequently misread metrics in SEO — adding new long-tail queries at low ranks can pull average position down even while primary rankings improve.

B

• Backlink

  A backlink is an HTML link from one domain pointing to a page on another domain. It is the atomic unit of off-page SEO and one of Google's three named top ranking signals. The `rel` attribute (none, nofollow, sponsored, ugc) determines how Google treats the link's authority transfer.

• BERT algorithm (BERT)

  BERT stands for Bidirectional Encoder Representations from Transformers. It is the language model Google rolled out to its search ranking and snippet systems in October 2019. BERT improved Google's understanding of prepositions, conjunctions, and overall query context — especially for conversational queries.

• Bing Copilot

  Microsoft Copilot in Bing is Microsoft's AI answer engine inside Bing Search. Launched February 7, 2023 as "Bing Chat" using GPT-4-class models with Bing-index grounding. Rebranded to "Copilot" on September 21, 2023 to align with Microsoft's broader Copilot product family.

• Bing Places

  Bing Places for Business is Microsoft's free local-listing platform — the equivalent of Google Business Profile for the Bing ecosystem. A claimed listing drives appearance in Bing Maps, Bing Search local answers, and (partially) Copilot's local recommendations. Often overlooked; meaningful where Bing has share.

• Black-hat SEO

  Black-hat SEO is search optimization that violates Google's Webmaster Guidelines. Tactics like cloaking, link buying, scraped content, doorway pages, and PBNs. High short-term leverage and high penalty risk. Reserved for high-burn-rate operations that can absorb domain loss.

• Brand visibility in LLMs

  Brand visibility in LLMs is the practice of tracking when, where, and how your brand appears inside model-generated answers — named, attributed, or described in context. It's distinct from search rankings and is now the core measurement category for generative engine optimization.

• Branded keyword

  A branded keyword is any query containing your brand name — "nike", "nike air max", "nike returns policy". Highest CTR of any keyword type (often 30-50% or more on the top result), highest conversion rate, and the most contested space when competitors run brand-bidding ads.

• Breadcrumbs

  Breadcrumbs are the navigational trail showing a page's position in a site hierarchy (Home > Category > Subcategory > Page). When marked up with BreadcrumbList schema, Google replaces the raw URL in the SERP with the breadcrumb path — a clearer, more clickable result for users.

• Broken link building

  Broken link building is a white-hat tactic: find a 404 link on a target page, contact the editor, and suggest your equivalent resource as the replacement. The pitch helps the editor (a real fix) and earns you a contextual backlink. Hit rates between 5% and 15% are typical.

• Bytespider

  Bytespider is the web crawler operated by ByteDance, the parent company of TikTok, used to gather training data for ByteDance's models including the Doubao family. It's known for aggressive request rates, frequently appears at the top of Cloudflare's "most-blocked AI bots" reports, and respects robots.txt when properly addressed.

C

• Canonical chain

  A canonical chain occurs when URL A canonicalizes to B, and B canonicalizes to C. Google may follow chains for a few hops, but treats long chains as signal noise and may pick a different canonical entirely. Always point each canonical directly at the final intended destination.

• Canonical tag

  A canonical tag (`<link rel="canonical" href="...">`) tells search engines which URL is the preferred version when near-duplicate content exists on multiple URLs. It is a hint, not a directive: Google may override it when its own signals point elsewhere.

• Canonicalization

  Canonicalization is the process by which Google selects one URL from a set of duplicate or near-duplicate URLs to represent the cluster in search results. The chosen URL is the canonical; the rest are alternates. Google weighs multiple signals — rel=canonical, redirects, sitemaps, internal links — to decide.

• ccTLD vs gTLD

  A ccTLD is a country-code top-level domain (.es, .de, .fr, .co.uk) and carries strong implicit geotargeting to that country. A gTLD is a generic top-level domain (.com, .net, .org, .io) and is global by default. The choice between them is one of the largest architectural decisions in international SEO.

• Citation rate

  Citation rate is the percentage of AI-generated answers (Perplexity, ChatGPT Search, AI Overviews, Gemini) that cite your domain across a defined query set. It's the GEO equivalent of share-of-voice for keyword rankings — and the emerging benchmark for how visible you are inside model outputs.

• ClaudeBot

  ClaudeBot is Anthropic's web crawler, identified by the user-agent `ClaudeBot`. It's used to gather public content for training Claude models. Block it via robots.txt with `User-agent: ClaudeBot`. Two earlier user-agents, `anthropic-ai` and `Claude-Web`, are deprecated but worth listing for safety.

• Click-through rate (CTR)

  Click-through rate is clicks divided by impressions, expressed as a percentage. In SEO, CTR is the leverage metric: a one-point improvement applied across millions of impressions compounds faster than gaining new rankings. Position-CTR benchmark curves are public estimates, not laws.

• Co-citation

  Co-citation occurs when two URLs are cited (linked) together by a third source, even without a direct link between them. The shared source acts as an implicit relevance signal. A concept Google has explored since the early 2010s, drawn from academic citation analysis.

• Co-occurrence

  Co-occurrence is when a brand or term appears alongside specific topic keywords across the web, without requiring a hyperlink between the mention and the brand's site. An implicit relevance signal that helps search engines associate entities with topics and intent.

• Common Crawl

  Common Crawl is a non-profit open-web archive that has crawled the public internet monthly since 2008. Its crawler is `CCBot`, and the resulting petabytes of HTML and text are the foundation of most public LLM training datasets, including GPT-3, LLaMA, and many others. Hosted at commoncrawl.org.

• Content brief

  A content brief is the structured pre-writing document for a single piece of content. It specifies the primary keyword, secondary keywords, search intent, target length, outline, internal links, external sources, and schema. The instrument that enforces editorial discipline at scale.

• Content gap analysis

  Content gap analysis is the process of identifying topics or keywords your competitors rank for that you don't yet cover. The output is a prioritized list of content briefs. Tooling: Ahrefs Content Gap, Semrush Topic Research, or GSC query exports cross-referenced against competitor visibility data.

• Content pruning

  Content pruning is the deliberate removal, consolidation, or noindexing of low-quality, thin, or outdated pages to lift site-wide quality signals. The counterintuitive lever: deleting pages can raise rankings on the survivors, because Panda and the helpful content system score sites in aggregate.

• Content refresh

  Content refresh is the systematic update of existing published pages — new data, new examples, expanded coverage, removed stale claims. Often higher ROI than publishing new content because the page already has age, links, and crawl history. Always update `dateModified` to signal the change.

• Content silos

  Content silos are tightly-linked clusters of content on a single topic, kept organizationally and link-wise separate from other clusters on the site. The 2010-era predecessor to the pillar-cluster model. Concentrates topical authority by giving Google an unambiguous read on what each section is about.

• Conversational Search

  Conversational search is multi-turn search where context from earlier queries persists into later ones. Instead of issuing a fresh query for each refinement, the user asks follow-ups like a chat. Standard in ChatGPT Search, Perplexity, Google AI Mode, and Bing Copilot.

• Conversion rate

  Conversion rate is the percentage of sessions or users that complete a defined goal — purchase, sign-up, lead form. The denominator choice (sessions vs users) shifts the number meaningfully. For SEO, the useful version is segmented by landing page and acquisition channel, not the sitewide aggregate.

• Core update

  A core update is a broad algorithmic refresh Google rolls out roughly three to four times a year, affecting rankings across all queries and verticals. There is no single fix for ranking drops — recovery is rooted in long-term content quality, expertise signals, and overall site trustworthiness.

• Core Web Vitals (CWV)

  Core Web Vitals (CWV) are three field-measured metrics — Largest Contentful Paint (LCP), Interaction to Next Paint (INP), and Cumulative Layout Shift (CLS) — that Google uses as signals of real user experience. A page passes CWV when all three are in the "good" band at the 75th percentile.

• Cost per click (CPC)

  Cost per click (CPC) is the average price advertisers pay for a click on Google Ads for a given keyword. For organic SEO, CPC is the cleanest public proxy for commercial intent: advertisers only sustain high bids when the keyword converts. High CPC roughly maps to high purchase-readiness.

• Crawl budget

  Crawl budget is the combination of crawl capacity (how much Googlebot can crawl without overloading a server) and crawl demand (how much it wants to). Under about 10,000 URLs it rarely matters; above that threshold, it decides which pages ever reach the index.

• Crawl trap

  A crawl trap is any URL pattern that generates an unbounded or pathologically large set of URLs. Common sources: infinite calendar archives, faceted navigation combinatorics, session IDs in query strings, and recursive relative-link bugs. Crawl traps drain crawl budget and frustrate indexing of pages that actually matter.

• Crawlability

  Crawlability is whether a URL can be discovered and fetched by a search engine crawler. It depends on internal linking exposure, robots.txt rules, DNS and server availability, and HTTP response codes. A URL can be crawlable without being indexable — and the reverse is also true.

• CrUX Report (CrUX)

  The Chrome User Experience Report (CrUX) is Google's public dataset of real-user performance metrics, collected from opt-in Chrome users on public websites. Powers the field data in PageSpeed Insights, the Core Web Vitals report in Search Console, and the basis for ranking signals. 28-day rolling window.

• Cumulative Layout Shift (CLS)

  Cumulative Layout Shift (CLS) measures the sum of unexpected layout shifts during a page visit. Each shift is scored by impact fraction × distance fraction. A page passes CLS when the total is ≤ 0.1 at the 75th percentile of real-user visits.

D

• Digital PR

  Digital PR is public-relations work aimed primarily at earning editorial backlinks and brand mentions in digital publications. It overlaps with traditional PR in tactics (pitches, data stories, expert sourcing) but measures success in linking root domains and share of voice, not column inches.

• Disavow file

  A disavow file is a plain-text list of referring URLs or domains submitted through Google Search Console's disavow tool, telling Google to ignore those backlinks when evaluating your site. After Penguin 4.0 (Sept 2016), most spam links are auto-neutralized, making disavow an edge-case tool.

• Discover feed

  Google Discover is the personalized content feed shown on the Google mobile app and Android home screen, formerly called Google Now. It serves articles algorithmically based on user interests — no query involved. Eligibility appears in Search Console only after Discover starts driving traffic.

• Dofollow link

  Dofollow is colloquial SEO shorthand for any HTML link that does not carry a `rel="nofollow"`, `rel="sponsored"`, or `rel="ugc"` attribute. It is the default link state — no markup is required. Such a link passes link equity and anchor-text relevance to the target URL.

• Domain Authority (DA)

  Domain Authority is a Moz proprietary score from 1 to 100 that predicts how well a domain may rank on Google. It is not a Google metric and Google does not use it. Treat it as a relative comparison tool between competitors, never as an absolute KPI to chase.

• Domain Rating (DR)

  Domain Rating is Ahrefs' proprietary 0-100 score reflecting the strength of a domain's backlink profile. It uses a logarithmic scale based on the quantity and quality of unique referring domains. Like Moz's DA, it is a vendor heuristic — Google does not use it.

• Doorway pages

  Doorway pages are pages or sites built primarily for search engines, funneling users to a single destination once they land. Classic patterns: city-by-city landing pages with near-identical content, multiple domains targeting close variants of the same query. A long-standing Google spam policy, tightened by the 2015 doorway update.

• Duplicate content

  Duplicate content is substantially identical content appearing on two or more URLs, either on the same site (internal) or across sites (cross-domain). There is no duplicate-content penalty — Google deduplicates by picking one URL as canonical and demoting the rest. Canonicals steer the choice.

E

• E-E-A-T (E-E-A-T)

  E-E-A-T stands for Experience, Expertise, Authoritativeness, Trustworthiness. It's the framework Google's Search Quality Raters use to evaluate content, added "Experience" in December 2022. Not a direct ranking score — it informs the quality signals that rankers learn from. Critical for YMYL topics.

• Ecommerce faceted navigation

  Ecommerce faceted navigation is the filter system on PLPs — color, size, brand, price, material — that lets users narrow product lists. Each filter combination is a URL. Without rules, the combinatorics generate millions of near-duplicate pages and a textbook crawl trap.

• Embedding

  An embedding is a fixed-length numeric vector that represents a piece of text (or image, audio, code) inside a learned semantic space, where similar meanings sit close together. Typical dimensions are 768, 1536, or 3072. Embeddings are the atomic data unit behind semantic search, RAG, and recommendation systems.

• Engagement rate (GA4) (GA4)

  Engagement rate in GA4 is the percentage of sessions classified as engaged. An engaged session lasts longer than 10 seconds, fires a conversion event, or records at least 2 page or screen views. The metric replaces Universal Analytics' bounce rate and is its mathematical inverse — almost.

• Entity SEO

  Entity SEO is optimizing for *entities* — people, places, organizations, products, concepts — in addition to keywords. Entities are how Google's Knowledge Graph and how LLM-driven answer engines understand the world. The shift from "ranking keywords" to "being the recognized source on a topic".

• ETag

  ETag is an HTTP response header containing an opaque entity-tag identifying a specific version of a resource. On subsequent requests, the client sends `If-None-Match: <etag>` and the server returns 304 Not Modified if unchanged, skipping the body. Googlebot honors this and saves bandwidth on stable assets.

• Evergreen content

  Evergreen content covers topics where search demand stays roughly stable across years. Examples: "how to write a CV", "what is HTTPS", "how to boil an egg". Distinguished from trending (one news cycle) and seasonal (annual spike). The compounding asset class of an SEO content strategy.

F

• Faceted navigation

  Faceted navigation is the filter and sort UI common on category and search pages — color, size, brand, price range. Each combination generates a URL. On large sites these combinations explode into millions of low-value URLs that drain crawl budget, dilute internal links, and create duplicate-content signals.

• Featured snippet

  A featured snippet is an answer extracted from a top-10 organic result and displayed at the top of the SERP, often called "Position 0". Four primary formats exist: paragraph, numbered or bulleted list, table, and video. Eligibility requires ranking in the top ten and matching question intent.

• First Contentful Paint (FCP)

  First Contentful Paint (FCP) is the moment the browser renders the first piece of DOM content — text, image, non-blank canvas, or SVG. Anything that proves to the user "the page is loading". Good under 1.8s, poor above 3s. Diagnostic, not a Core Web Vital.

G

• GA4 sessions (GA4)

  A GA4 session is a group of user interactions on your site within a time window. The default 30-minute inactivity timeout still applies, but unlike Universal Analytics, a campaign or source change mid-session no longer starts a new session. Session count is derived from `session_start` events.

• Gemini Search

  Gemini Search refers to the web-grounded answer mode inside Google Gemini (gemini.google.com), where the assistant uses the Google Search grounding tool to fetch live results and cite them. Distinct from AI Overviews (inside google.com SERPs) and AI Mode (the AI-only Search tab).

• Generative Engine Optimization (GEO)

  Generative Engine Optimization (GEO) is the practice of optimizing content so large language model answer engines — ChatGPT, Perplexity, Google AI Overviews, Bing Copilot — cite, quote, and reference your brand inside generated responses. The term was coined in a 2023 paper by Aggarwal et al. (arXiv:2311.09735).

• Geo-modified keyword

  A geo-modified keyword is a search query that explicitly includes a location ("plumber Madrid", "dentist near me", "best ramen Brooklyn"). These queries trigger local-intent ranking systems — the Local Pack, Maps results, and city-tagged organic listings — and behave differently from generic queries.

• Geotargeting

  Geotargeting is the practice of telling Google that a site or a section of it is intended for users in a specific country. The signals are layered: ccTLD (implicit), hreflang (page-level), server/IP location (weak), and Search Console country targeting (deprecated for top-level since 2022, now algorithmic).

• Google Business Profile (GBP)

  Google Business Profile (GBP) is Google's free listing product for businesses, renamed from Google My Business in November 2021. It controls how a business appears in Google Search, Maps, and the Local Pack. Verification of the profile is the prerequisite for any local visibility on Google.

• Google Merchant Center

  Google Merchant Center is the platform where retailers upload product feeds to power Shopping ads, free Shopping listings, the Shopping tab, and product-related surfaces in Search. Distinct from on-page Product schema — Merchant Center reads a separate structured feed.

• Google-Extended

  Google-Extended is an opt-out token for `robots.txt` that blocks your content from training Google's generative models (Gemini, Vertex AI generative APIs) without affecting Googlebot, Google Search ranking, or indexing. Introduced September 2023 as a separate dial from search inclusion.

• GPTBot

  GPTBot is OpenAI's web crawler, identified by the user-agent token `GPTBot` and used to gather public content for training future models. Blocking it in robots.txt prevents training use but does not affect ChatGPT Search, which uses a separate user-agent (`ChatGPT-User`) for live retrieval.

• Guest posting

  Guest posting is publishing content on another site, usually with an author bio and a backlink to your own site. Legitimate when the post is topical, original, and editorially driven. Considered a link scheme by Google when industrialized — large-scale, keyword-anchored, paid-for placements.

H

• HARO (Help A Reporter Out) (HARO)

  HARO ("Help A Reporter Out") was a daily email service connecting journalists with expert sources for stories. Cision rebranded it as Connectively in 2024. SEO teams use it to earn high-authority citations and editorial backlinks by responding to source requests in their domain.

• Heading hierarchy

  Heading hierarchy is the logical nesting of H1-H6 tags in a document. One H1 per page that names the topic, H2s for major sections, H3s nested inside H2s for subsections. It signals document structure to crawlers, screen readers, and the LLM scrapers now extracting content for AI answers.

• Helpful Content System (HCU)

  The Helpful Content System is Google's framework for distinguishing people-first content from search-engine-first content. Originally launched in August 2022 as a separate sitewide classifier, it was folded into the core ranking signals during the March 2024 core update.

• Hreflang

  Hreflang is an annotation that tells search engines which language or regional version of a page to serve to which user. It can be delivered via `<link>` tags, HTTP headers, or XML sitemaps. Bidirectional references are mandatory — each version must reference all the others.

• HSTS (HSTS)

  HSTS (HTTP Strict Transport Security) is a response header that tells browsers to load a domain only over HTTPS for a declared duration. Defined in RFC 6797. It eliminates the HTTP-to-HTTPS redirect on repeat visits and blocks downgrade attacks. Sites can submit to the HSTS preload list for first-visit protection.

• HTTP 301 redirect

  A 301 is the HTTP "Moved Permanently" status code. It signals that a URL has changed location for good and clients should update bookmarks and references. Since 2016 Google has treated 301s as passing effectively all link equity to the target. The default redirect for permanent moves.

• HTTP 302 redirect

  A 302 is the HTTP "Found" status — a temporary redirect. It tells clients the original URL is the canonical one and they should keep using it. Google historically treated 302s differently from 301s, but since 2016 a long-lived 302 is consolidated to the target like a 301. Use only for true temporary moves.

• HTTP 308 redirect

  A 308 is the HTTP "Permanent Redirect" status code, defined in RFC 7538 (2015). It behaves like a 301 — permanent move, full link equity passed — except it preserves the original HTTP method. A POST stays a POST. The right choice for permanent redirects on APIs and form endpoints.

• HTTP 404 error

  A 404 is the HTTP "Not Found" status code. It tells clients the URL doesn't exist on this server right now and may or may not return. Google treats 404 as a soft signal — pages are eventually deindexed, but slowly. Healthy sites have some 404s; the problem is patterns, not individual hits.

• HTTP 410 Gone

  A 410 is the HTTP "Gone" status code — an explicit "this URL was here, has been intentionally removed, and is not coming back". Stronger than 404, which is ambiguous about future availability. Google has confirmed it deindexes 410s faster than 404s in practice.

• HTTP 503 error

  A 503 is the HTTP "Service Unavailable" status code. It tells clients the server is up but cannot handle the request right now and to try again later. Paired with a Retry-After header, it's the SEO-safe maintenance signal — Googlebot pauses crawling without deindexing.

• HTTP status codes

  HTTP status codes are three-digit responses servers return for every request, grouped into five classes: 1xx informational, 2xx success, 3xx redirection, 4xx client errors, 5xx server errors. Googlebot reads them on every fetch to decide whether to index, recrawl, drop, or back off.

• Hummingbird

  Hummingbird is the algorithm overhaul Google announced in September 2013 — the largest rewrite since Caffeine in 2010. It shifted ranking from keyword-by-keyword matching toward parsing the full query for intent and meaning. Hummingbird is the foundation later systems like RankBrain, BERT, and MUM extended.

I

• Image pack

  An image pack is an inline horizontal carousel of image results displayed on the SERP for queries with visual intent. The pack typically shows 4-12 thumbnails. Clicking any thumbnail opens Google Images, not the source URL — a click-path detour the publisher rarely recovers.

• Image SEO

  Image SEO is the practice of making images findable in Google Image Search, in image packs on regular SERPs, and as social-share previews — while keeping page weight low. It covers filename, alt text, dimensions, format, responsive delivery, lazy loading, and the image sitemap.

• Impressions (GSC) (GSC)

  Impressions in Google Search Console count how many times a URL from your site appeared on a search results page for any user, for any query. Counted once per result per SERP regardless of whether the user scrolled to it. Below Google's anonymization threshold the query is hidden but the impression still shows.

• Indexability

  Indexability is whether a crawlable URL is eligible to appear in a search engine's index. Blocked by noindex directives, canonical tags pointing elsewhere, thin or duplicate content, or quality filters. Crawlability is a prerequisite but not a guarantee.

• IndexNow

  IndexNow is an open push-based indexing protocol launched by Microsoft and Yandex in October 2021. Sites publish a key file at the root and POST changed URLs to IndexNow endpoints; participating search engines fetch them on notification rather than waiting for the next crawl cycle.

• Infinite scroll SEO

  Infinite scroll SEO is the problem of making continuously-loading list pages crawlable. Googlebot doesn't scroll — it requests URLs. A pure JS infinite scroll where new items load on scroll without changing the URL is invisible to crawlers beyond the first page. The fix: paginated alternative URLs.

• Information gain

  Information gain is a score described in Google's [US Patent 11620273](https://patents.google.com/patent/US11620273B2/) measuring how much new information a given document adds beyond what the user has already seen in other documents on the same topic. Rewards original substantive contribution; penalizes the third article on a SERP that just restates the first two.

• Interaction to Next Paint (INP)

  Interaction to Next Paint (INP) measures the worst interaction latency during a page visit — time from a user's tap, click, or keypress until the browser paints the next frame. Replaced FID as a Core Web Vital in March 2024. Good: ≤ 200 ms at the 75th percentile.

• Internal linking

  Internal linking is the practice of linking from one page on your site to another. It distributes link equity (PageRank), signals which pages you consider most important, and helps crawlers discover content. The one ranking lever you control completely — no outreach, no negotiation required.

• Intrusive interstitial

  An intrusive interstitial is a modal that blocks the primary content of a mobile page right after the user arrives from search. Google has used it as a ranking signal since January 2017. Cookie banners, age verification, and login walls for paywalled content are explicitly exempt from the penalty.

J

• JavaScript SEO

  JavaScript SEO is the practice of making JS-rendered content discoverable, crawlable, and indexable by search engines. It covers server-side rendering, client-side rendering, hydration, dynamic rendering, and the trade-offs each rendering strategy creates for crawlers like Googlebot, Bingbot, and AI bots.

• JSON-LD

  JSON-LD (JSON for Linked Data) is a JSON-based format for machine-readable annotations, standardized by W3C in 2014. In SEO it's the preferred way to deliver Schema.org structured data — a `<script type="application/ld+json">` block, typically in the HTML `<head>`, decoupled from the visible DOM.

K

• Keyword cannibalization

  Keyword cannibalization is when two or more pages on your site target the same query with overlapping intent, so Google can't decide which to rank. Ranking signals split between them and none reaches its full potential. Detected in GSC when multiple URLs alternate for one query.

• Keyword difficulty (KD)

  Keyword difficulty (KD) is a tool-specific score from 0 to 100 estimating how hard it would be to rank organically on the first page for a keyword. Each tool calculates it differently, so the same keyword can show KD 28 in one tool and KD 64 in another. Use it directionally, never as a hard threshold.

• Knowledge Graph

  The Knowledge Graph is Google's structured database of entities — people, places, organizations, things — and the relationships between them. Launched in 2012, it powers knowledge panels, the entity disambiguation layer in ranking, and increasingly the grounding for AI Overviews.

• Knowledge panel

  A knowledge panel is the entity card Google displays on the right side of desktop SERPs (or top on mobile) for queries about a specific person, place, organization, product, or thing. Powered by the Knowledge Graph, it aggregates facts from Wikipedia, Wikidata, structured data, and licensed feeds.

• KPI tree

  A KPI tree is a hierarchical structure linking a business outcome metric at the top (revenue, profit) to the leading indicators that drive it (conversions, sessions, impressions, rankings, technical health). It is the discipline that prevents teams from reporting metrics no decision depends on.

L

• Lab vs Field Data

  Lab data is synthetic — a single page load under controlled, throttled conditions, repeatable run-to-run. Field data is real users — millions of loads under varied devices, networks, and locations. Google ranks on field data (CrUX); you optimize against lab (Lighthouse) because it's diagnostic and reproducible.

• Language targeting

  Language targeting is hreflang configuration that addresses users by language alone — `en`, `es`, `de`, `fr` — without specifying a region. Use it when one language version serves all regions of that language. It's simpler than locale targeting and right when the content genuinely doesn't vary by country.

• Largest Contentful Paint (LCP)

  Largest Contentful Paint (LCP) is the time from navigation start until the largest image or text block in the initial viewport renders. It's one of Google's three Core Web Vitals. Good: ≤ 2.5 seconds at the 75th percentile of real-user visits.

• Last-Modified header

  Last-Modified is an HTTP response header containing the date a resource last changed (RFC 7232 format). On subsequent requests, clients send `If-Modified-Since` with that date; the server returns 304 Not Modified if nothing has changed. Googlebot honors this and saves crawl capacity on stable URLs.

• Lighthouse

  Lighthouse is Google's open-source auditing tool for web pages. It runs five category audits — Performance, Accessibility, Best Practices, SEO, and PWA — and produces scores from 0-100 plus actionable diagnostics. Generates lab data, not field data. Available in Chrome DevTools, CLI, Node, and PageSpeed Insights.

• Link equity

  Link equity is the ranking authority transmitted from one page to another through hyperlinks. It's the modern term for what was once called PageRank flow or link juice. Each outbound link on a page splits its available equity, so equity dilutes as the number of outbound links grows.

• Link rot

  Link rot is the gradual decay of links as pages move, sites go offline, or domains expire. It hits both your inbound link profile (links to you that stop working) and your outbound graph (links from your site that lead to 404s). A continuous loss SEO programs need to monitor and replace.

• Link velocity

  Link velocity is the rate at which a site acquires (or loses) backlinks over time. Sudden, unexplained spikes are flagged by spam filters as likely paid or programmatic patterns. Steady, varied growth signals natural editorial earning. The shape matters more than the absolute count.

• Linking root domain

  A linking root domain is a unique referring domain pointing at your site, regardless of how many individual pages on that domain link to you. Three different articles from the same publisher count as one linking root domain. The headline trust metric in most backlink tools.

• LLM Grounding

  LLM grounding is the practice of constraining a language model's output to retrieved sources or structured data, rather than letting it generate from parametric memory alone. Implemented via RAG, tool use, or system prompts that require citations. The mechanism behind inline citations in AI answers.

• llms.txt

  llms.txt is a proposed standard introduced by Jeremy Howard (Answer.AI) in September 2024: a Markdown file at `/llms.txt` that gives LLMs a clean, prioritized index of a site's most useful content. Adoption as of early 2026 is real but not universal; spec lives at llmstxt.org.

• Local citation

  A local citation is any web mention of a business's name, address, and phone (NAP), whether or not it includes a link. Citations live on directories like Yelp, BBB, Apple Maps, Yellow Pages, and industry-specific sites. They're the off-site backbone of local SEO — Google reads them as identity signals.

• Local landing page

  A local landing page is a dedicated page targeting a single location/service combination — "Plumbing in Móstoles", "Dentist Brooklyn Heights". It's the unit of multi-location SEO: NAP, hours, embedded map, location-specific content, schema markup, and reviews tied to that one location.

• Local pack

  The local pack — also called the map pack or 3-pack — is the SERP block showing a map with three local business listings, displayed for queries with local intent. It pulls data from Google Business Profile and ranks by relevance, distance, and prominence.

• Locale targeting

  Locale targeting is the practice of targeting a specific language + region combination — `en-US`, `en-GB`, `es-MX`, `es-ES`, `pt-BR`, `pt-PT`. It's the canonical unit [hreflang](/blog/hreflang) operates on and the right granularity when content varies by region within a language.

• Log file analysis

  Log file analysis is the parsing of web server access logs to see exactly which URLs search crawlers (Googlebot, Bingbot, others) have requested, at what rate, and with what response codes. It's the only ground-truth source for what crawlers actually do on your site.

• Long-tail keyword

  A long-tail keyword is a multi-word, lower-volume query with specific intent — for example "best running shoes for flat feet under 100 dollars" instead of "running shoes". Easier to rank for, higher conversion intent, individually small. The tail collectively beats the head on most ecommerce and B2B sites.

M

• Meta description

  The meta description is the `<meta name="description" content="...">` tag in a page's `<head>`. Search engines use it as a candidate for the SERP snippet. Google rewrites roughly 70% of them based on the user's query. Target length: 150-160 characters, action-oriented and specific.

• Mixed content

  Mixed content is an HTTPS page that loads subresources — images, scripts, stylesheets, iframes — over plain HTTP. Browsers block active mixed content by default since Chrome 80 (Feb 2020) and progressively block passive mixed content too. It breaks the HTTPS lock icon and degrades the security signal Google has used in ranking since 2014.

• Mobile-first indexing

  Mobile-first indexing means Google primarily uses the mobile version of a page for indexing and ranking. Announced 2016, default for new sites since July 2019, full rollout completed in 2023. Content, structured data, and links must exist on the mobile version or Google won't see them.

• Multi-location SEO

  Multi-location SEO is the practice of optimizing an organic and local presence for businesses with many physical locations — chains, franchises, multi-clinic providers, dealer networks. It covers site architecture, GBP management, content uniqueness, citation governance, and reporting at scale.

• MUM algorithm (MUM)

  MUM stands for Multitask Unified Model, the language model Google announced in May 2021 as the successor to BERT. Google's claim: 1000x more capable than BERT, multimodal (text plus image), and multilingual across 75 languages. MUM powers Things to Know and underpins parts of AI Overviews.

N

• NAP consistency (NAP)

  NAP stands for Name, Address, Phone — the three identity fields that must match across a business's Google Business Profile, website, and every citation on the web. Discrepancies (abbreviations, old phone numbers, suite formatting) degrade the trust signal Google uses to confirm a real business.

• Nofollow

  Nofollow is a link-level attribute (`rel="nofollow"`) telling search engines this link should not pass ranking credit. Since 2019 Google treats it as a hint, not a directive — reserving the right to ignore it. Sibling attributes for more specific cases: `rel="ugc"` and `rel="sponsored"`.

• Noindex

  Noindex is a directive telling search engines not to include a URL in their index. Delivered via `<meta name="robots" content="noindex">` or an `X-Robots-Tag: noindex` HTTP header. Requires the page to be crawlable — a URL blocked by robots.txt can never be seen as noindex.

O

• Organic traffic

  Organic traffic is the count of visits arriving on your site through unpaid search engine results. It excludes paid search (Google Ads), referrals from other sites, direct visits, and social. In GA4 it maps to the "Organic Search" channel grouping, with source identification driven by referrer parsing.

• Organic visibility

  Organic visibility is a composite metric estimating your overall SERP presence — the count of keywords you rank for, weighted by position and search volume. Each tool computes it differently. Useful as a directional trend signal, less useful as an absolute number.

• Orphan page

  An orphan page is a URL on your site with no internal links pointing to it. It may still be crawlable via sitemap entries or backlinks, but receives no internal PageRank flow. Search engines crawl orphans rarely and index them reluctantly — often a symptom of unfinished migrations.

• Out-of-stock SEO

  Out-of-stock SEO is how a site handles product URLs when stock runs out: temporarily unavailable, seasonally gone, or permanently discontinued. The right response depends on which of those three states applies. The wrong response — 404 everything, or redirect every dead PDP to the home — destroys ranking equity built up over years.

P

• Page authority (PA)

  Page authority is a URL-level score estimating how likely a single URL is to rank in search results. Moz coined the metric as a 1-100 logarithmic scale. Google uses its own internal signals, not Moz's; PA is a correlated proxy, not ground truth. High PA does not guarantee ranking.

• Page experience update

  The page experience update rolled out to mobile in June 2021 and to desktop in February 2022. It bundled Core Web Vitals (LCP, FID/INP, CLS) with HTTPS, no intrusive interstitials, and mobile-friendliness into a unified ranking signal. Now described as part of Google's broader page experience ranking systems rather than a discrete update.

• PageRank

  PageRank is the original Google ranking algorithm, patented by Larry Page and Sergey Brin at Stanford in 1998. It scores pages by the quantity and quality of inbound links. The public Toolbar PageRank score was deprecated in 2016, but Google has confirmed the underlying signal is still operational.

• PageSpeed Insights (PSI)

  PageSpeed Insights (PSI) is Google's web tool combining two data sources per URL: Lighthouse lab data (one synthetic run) and CrUX field data (28-day rolling real-user data from opt-in Chrome users). The default public reference for "is this page fast enough" and Core Web Vitals pass/fail.

• Pagination SEO

  Pagination SEO is the handling of paginated series — category pages, archives, search results — split across `/page/2`, `/page/3`, etc. Google deprecated `rel=prev/next` for indexing in 2019. Each page should self-canonical. View-all is an option only when the full list is reasonably small.

• Panda algorithm

  Panda was the algorithmic update Google launched on February 23, 2011, to target thin, duplicate, and low-quality content — particularly content farms that had dominated SERPs in 2010. Periodic refreshes ran for five years, then Panda was folded into Google's core algorithm in January 2016.

• Parameter handling

  Parameter handling is how Google decides which URL parameters change content meaningfully versus which are noise. Sort orders, filters, session IDs, tracking codes, and pagination all use parameters. The GSC parameter tool was deprecated in April 2022 — Google now relies on its own algorithms.

• Parasite SEO

  Parasite SEO is the practice of publishing content on a high-authority host domain to rank queries the publisher's own site couldn't reach. Common hosts: major news sites, LinkedIn articles, Medium, university subdomains. Google targeted the publisher-host variant directly with the site reputation abuse policy in May 2024.

• Penguin algorithm

  Penguin was Google's algorithmic update launched on April 24, 2012, to combat manipulative backlink patterns — link schemes, exact-match anchor abuse, paid links, low-quality directories. Penguin 4.0 (September 2016) made it real-time and granular, devaluing bad links rather than penalizing whole sites.

• People Also Ask (PAA)

  People Also Ask (PAA) is an expandable question box on the SERP showing related questions with collapsible answers. Each click loads more questions dynamically, sometimes infinitely. Sources are top-ranking pages — eligibility is roughly the same as featured snippets.

• People Also Search For

  People Also Search For is a related-queries box that appears under a search result when a user clicks through and then returns to the SERP. The box shows alternative queries Google thinks better match the user's intent. Often called "pogo-stick" behavior because the user bounced.

• Perplexity

  Perplexity AI is an answer engine that pairs every generated claim with numbered inline citations to web sources. Founded 2022 (Aravind Srinivas). Free tier with usage limits, Pro tier with model choice and unlimited Pro Search, Spaces for shared collections, and the Sonar API for developers.

• PerplexityBot

  PerplexityBot is the indexing crawler used by Perplexity AI to build its search index. `Perplexity-User` is a separate user-agent that fetches pages on demand when a user asks a question. Distinguishing the two lets you opt out of indexing while remaining citable in live answers.

• Pillar-cluster model

  The pillar-cluster model is a content architecture: one broad "pillar" page on a topic, linked bidirectionally with many specific "cluster" pages covering subtopics. Replaces older content silos. The pattern Google's topical authority signal — and LLM grounding — most cleanly recognizes.

• Private blog network (PBN)

  A private blog network (PBN) is a network of seemingly independent sites built or acquired with the sole purpose of linking back to a target site the operator wants to rank. An explicit violation of Google's link-spam guidelines, hit by Penguin in 2012 and continuously since.

• Product detail page (PDP)

  A product detail page (PDP) is the URL that represents one purchasable item. It carries the title, gallery, price, variant selector, buy button, schema, and review block. PDPs are where commercial intent converts and where most ecommerce ranking work concentrates.

• Product feed (GTIN/MPN)

  A product feed is the structured catalog file — XML, CSV, TSV, Google Sheet, or Content API payload — that an ecommerce site uploads to Google Merchant Center. It contains one row per SKU with attributes Google needs to render Shopping listings: title, image, price, availability, GTIN, brand.

• Product list page (PLP)

  A product list page (PLP) is the category or collection page that lists multiple products with filters and sort. It targets head-to-mid keywords ("running shoes", "women's leather jackets") and funnels link equity to the PDPs it contains.

• Product schema

  Product schema is the schema.org/Product structured-data type that describes a single sellable item — name, image, price, availability, ratings. Required for Google's product rich results in classic search and the Shopping tab. Implemented as JSON-LD inside the product detail page.

• Product variant SEO

  Product variant SEO is the decision of how to handle color, size, and SKU variations of one product: a single canonical PDP with selectors and schema hasVariant, or separate URLs per variant. The default pattern shifted in October 2023 when Google released hasVariant support.

Q

• Quality Rater Guidelines (QRG)

  The Quality Rater Guidelines is the public document Google publishes for the thousands of human contractors who evaluate search results. It defines E-E-A-T, YMYL, page quality scoring, and "needs met" assessments. Not the algorithm itself — but the training signal that shapes ranking systems.

• Question keyword

  A question keyword is a query phrased as a question — starting with who, what, when, where, why, how, can, does, is, should. These trigger featured snippets, populate "People Also Ask" boxes, and feed AI overviews. The cheapest research input for any content brief.

R

• RankBrain

  RankBrain is Google's first major machine learning system applied to search ranking, confirmed in October 2015. It helps interpret novel, ambiguous, or never-before-seen queries by mapping them to similar known queries. Still operational today as one of many signals in the ranking system.

• Real User Monitoring (RUM)

  Real User Monitoring (RUM) is the practice of measuring performance and Core Web Vitals from actual user sessions rather than synthetic lab tests. CrUX is the public RUM dataset Google uses for ranking; private RUM tools (SpeedCurve, DebugBear, Datadog, New Relic) add per-session granularity, custom dimensions, and authenticated-page coverage.

• Reciprocal linking

  Reciprocal linking is the arrangement where two sites link to each other, often by explicit agreement. A common tactic in the early 2000s, flagged as a link scheme since the 2012 Penguin update when used at scale or via link-exchange schemes. Natural reciprocation between genuine partners is fine.

• Redirect chain

  A redirect chain is two or more sequential HTTP redirects — URL A → 301 → B → 301 → C. Each hop adds latency, consumes crawl budget, and risks clients stopping mid-chain. Most crawlers abandon the chain after 5-10 hops; users wait through every one.

• rel=canonical

  `rel="canonical"` is the HTML attribute declaring which URL is the preferred version among duplicates or near-duplicates. It's the technical mechanism behind the canonical-tag concept. Available as `<link rel="canonical">`, as an HTTP `Link` header, and implicitly via XML sitemap entries.

• Render-blocking Resources

  Render-blocking resources are CSS or JavaScript files in the document `<head>` that the browser must download and process before it can render anything. They delay First Contentful Paint and Largest Contentful Paint. Common fixes: inline critical CSS, async/defer JS, split CSS by media, preload the LCP element.

• Resource Hints

  Resource hints are HTML link directives that tell the browser to set up connections or fetch resources earlier than the parser would discover them. The five: dns-prefetch, preconnect, preload, prefetch, and modulepreload. Each targets a different stage of the loading pipeline. Used carelessly, they hurt — used surgically, they cut LCP.

• Retrieval-Augmented Generation (RAG)

  Retrieval-Augmented Generation (RAG) is the pattern of fetching relevant documents from a corpus, injecting them into the prompt as context, and generating an answer grounded in that retrieved content. Coined by Lewis et al. in [arXiv:2005.11401](https://arxiv.org/abs/2005.11401) (2020). Powers most cited AI answers in 2026.

• Review schema

  Review schema refers to the schema.org `Review` and `AggregateRating` types that mark up customer ratings in structured data. They power the star-rating rich result in Google. Heavily restricted in September 2019 to limit "self-serving" abuse — businesses can't mark up reviews they wrote about themselves.

• Reviews update

  The reviews update — now called the Reviews System and folded into the core ranking system in November 2023 — targets thin or aggregator-style review content. Google rewards first-hand evaluation, side-by-side comparison, original photography or video, and clear evidence the reviewer used the product.

• Rich result

  A rich result is a SERP listing visually enhanced by structured data — product prices, recipe ratings, event dates, FAQ expansions, and more. Eligibility depends on schema markup that matches a documented Google rich-result type. Test eligibility with the Rich Results Test.

• Robots.txt

  Robots.txt is a plain-text file at the root of a host (`/robots.txt`) that tells crawlers which paths they may or may not fetch. It controls crawling, not indexing — a blocked URL can still appear in search results if it has inbound links.

• ROI vs ROAS (ROI)

  ROI (Return on Investment) is profit divided by total cost, expressed as a percentage. ROAS (Return on Ad Spend) is revenue divided by ad spend, expressed as a multiple. The two are not interchangeable. SEO has no per-impression spend, so it uses adjusted-ROI frameworks tied to fixed program cost.

S

• Schema.org

  Schema.org is the shared vocabulary — types and properties — maintained by Google, Microsoft, Yahoo, and Yandex to describe entities on the web. Launched 2011. Machine-readable annotations using this vocabulary are what search engines parse into rich results.

• Search Generative Experience (SGE)

  Search Generative Experience (SGE) was Google's opt-in Labs preview of generative AI answers in Search, launched May 10, 2023 at Google I/O. It graduated to general availability as AI Overviews on May 14, 2024. SGE no longer exists as a separate product — references to it now point to AI Overviews.

• Search intent

  Search intent is the underlying goal a user has when issuing a query — what they want to learn, find, compare, or do. The four classical types are informational, navigational, commercial, and transactional. Matching intent dominates modern ranking; exact-keyword matching does not.

• Search volume

  Search volume is the estimated number of times users search a specific keyword on Google in a given month, usually averaged over the past 12 months. Every figure you see is an estimate. Sources disagree by 50% or more on the same keyword, and only Google has the ground truth.

• SearchGPT

  SearchGPT was OpenAI's prototype answer engine, announced July 25, 2024. It was integrated into ChatGPT as "ChatGPT Search" on October 31, 2024, and rolled out broadly (free tier included) on December 16, 2024. The standalone SearchGPT product no longer exists as a separate surface.

• Seasonal keyword

  A seasonal keyword sees demand spike at predictable calendar windows — "black friday deals", "back to school supplies", "halloween costumes", "christmas gift ideas". Annual pattern is sharp and repeatable. Plan publish dates 2-3 months ahead of the spike so content has time to rank.

• Semantic HTML

  Semantic HTML is the practice of using HTML elements that carry meaning about their content (`<article>`, `<section>`, `<nav>`, `<main>`, `<aside>`, `<header>`, `<footer>`) instead of generic `<div>` containers. Crawlers, screen readers, and LLM scrapers rely on these tags to extract structure.

• SERP analysis

  SERP analysis is examining the live search results page for a target keyword to infer search intent, expected content format, SERP-feature saturation, and competitive effort. Five minutes of manual review beats an hour of tool-only research and prevents the most expensive content mistakes.

• Service-area business (SAB)

  A service-area business (SAB) serves customers at the customer's location rather than at a public storefront — plumbers, mobile mechanics, locksmiths, cleaning services. On Google Business Profile, an SAB declares service areas by city or postal code instead of inviting walk-ins, and the address is optionally hidden.

• Share of voice (SoV)

  Share of voice in SEO is the proportion of total visibility you hold in a defined keyword universe relative to your competitors. The formula varies by tool: some weight by search volume, others by position-CTR estimates, most combine both. The keyword set chosen matters more than the formula.

• Shopping results

  Shopping results are the product listings on the SERP, mixing paid Product Shopping ads with organic shopping listings. Ads pull from a Google Merchant Center feed. Organic listings pull from the same feed plus product schema on indexed pages — opt-in to free listings is required.

• Site links

  Site links are the indented sub-page links Google shows beneath a top result, most often for branded or navigational queries. Up to 6-10 links can appear in expanded layouts. Generation is automatic, but you can influence which pages qualify and demote ones you don't want.

• Site reputation abuse

  Site reputation abuse is publishing third-party content on a high-authority host site primarily to manipulate rankings, when that content has little oversight from the host and is unrelated to its main purpose. Codified as a Google spam policy in May 2024. Targets coupon sections, sponsored deal hubs, and rented subdirectories.

• Sitemap index

  A sitemap index is an XML file that lists multiple individual sitemap files rather than URLs directly. Required when a single sitemap would exceed the limits of 50,000 URLs or 50 MB uncompressed. The mandatory pattern for any site whose URL count puts it past those caps.

• Soft 404

  A soft 404 is a URL that returns HTTP 200 but the content tells the user the page doesn't exist — empty product page, "no results found", a generic template with no real content. Google detects soft 404s heuristically and flags them in Search Console under "Page indexing".

• SPA SEO (SPA)

  SPA SEO is the practice of making Single-Page Applications (React, Vue, Angular, Svelte) discoverable and indexable. The rendering strategy is the primary decision: SSR, SSG, hybrid, or dynamic rendering. Pure client-side rendering forces Google's WRS into a second-wave fetch, slowing indexing.

• Spam update

  A spam update is a targeted enforcement run against Google's published spam policies — cloaking, sneaky redirects, scaled content abuse, link schemes, doorways. Distinct from a core update, which reweights quality signals broadly. Recovery requires removing the offending pattern, not waiting it out.

• Speed Index

  Speed Index is a lab metric that measures how quickly the contents of a page are visually populated during load. Calculated by sampling frames from a video of the load and integrating the percentage of visual completeness over time. Lower is better. Good under 3.4s, poor above 5.8s.

• Sponsored link

  A sponsored link is any outbound link marked with `rel="sponsored"`. Google introduced the attribute in September 2019 as the required way to disclose paid, affiliate, or sponsorship-driven links. Non-compliance — using unmarked dofollow links for paid placements — risks link-spam filtering.

• Structured data

  Structured data is machine-readable metadata embedded in a web page that describes its content in a standardized vocabulary (usually Schema.org). It enables rich results — stars, prices, events, FAQs — in the SERP and helps search engines understand the page's entities.

• Subdomain vs subdirectory

  The choice between hosting content at `blog.example.com` (subdomain) or `example.com/blog` (subdirectory). Google states it treats both equivalently. Practitioners report subdirectories consistently inherit site authority more directly. For most SEO-driven decisions, default to subdirectory.

T

• TF-IDF (TF-IDF)

  TF-IDF (Term Frequency - Inverse Document Frequency) is a classic information retrieval scoring formula. It ranks how important a term is in a document relative to a corpus. Modern Google ranking has moved far beyond TF-IDF into neural embeddings, but the concept persists in content optimization tools.

• Thin content

  Thin content is a page with insufficient substance to satisfy the topic it claims to cover. Typical examples: auto-generated directories, stub pages, category listings without intros, doorway pages. Google's quality filters demote or deindex them as "soft 404" or low-value rather than penalizing.

• Things to know

  Things to Know is a SERP feature that displays expandable subtopics for broad queries, introduced in 2022 and powered by Google's MUM algorithm. Each subtopic expands to show related aspects of the query. Many implementations are display-only — the subtopics aren't clickable.

• Time to First Byte (TTFB)

  Time to First Byte (TTFB) is the interval between the browser sending a request and receiving the first byte of the response. It captures DNS, TCP, TLS, request queueing, and server processing. Google's thresholds: good under 800ms, needs improvement 800-1800ms, poor above 1800ms. Strong upstream correlate of LCP.

• Time to Interactive (TTI)

  Time to Interactive (TTI) is the lab metric for when a page becomes reliably responsive — first FCP, then 5 seconds of network and main-thread quiet. Removed from Lighthouse's scoring weight in v10 (2023) and effectively deprecated as a public metric. INP replaced it for measuring real-user interactivity.

• Title tag

  The title tag (`<title>` in the HTML `<head>`) is the primary SERP headline and the browser tab text. Google rewrites roughly 60% of declared titles based on query intent. Target 50-60 characters, lead with the primary keyword, end with brand for clickthrough.

• TLS / SSL (TLS)

  TLS (Transport Layer Security) is the cryptographic protocol that secures HTTPS connections. SSL (Secure Sockets Layer) is the legacy term still used colloquially — actual SSL is deprecated; modern sites use TLS 1.2 or 1.3. HTTPS has been a confirmed (minor) Google ranking signal since August 2014.

• ToFu / MoFu / BoFu

  ToFu, MoFu, and BoFu stand for Top, Middle, and Bottom of Funnel. The labels describe where a user sits in the path from awareness to purchase. In SEO, the funnel stage maps to search intent: ToFu to informational, MoFu to commercial investigation, BoFu to transactional or navigational queries.

• Top stories

  Top Stories is the news carousel that appears on the SERP for queries with news intent or freshness signals. It rotates a handful of recent articles with publisher logos and timestamps. Eligibility requires inclusion in Google News and meeting technical and editorial signals.

• Topical authority

  Topical authority is a search engine's assessment of how thoroughly a site covers a subject. Built through related-article depth, entity co-occurrence, consistent internal linking within a cluster, and sustained publication cadence. Stronger topical authority lifts rankings across the entire cluster.

• Total Blocking Time (TBT)

  Total Blocking Time (TBT) is the sum, in milliseconds, of the portions of long tasks (>50ms) between FCP and TTI that exceed 50ms. Lab-only metric in Lighthouse. The closest synthetic proxy for INP, the field Core Web Vital for interactivity. Good under 200ms, poor above 600ms.

• Toxic link

  A toxic link is a vendor-defined label for an inbound link a tool considers potentially harmful — usually based on heuristics like spammy TLDs, no organic traffic, link-farm patterns. Google does not use the term and most toxic-flagged links don't actually require any action.

• Trailing slash

  The trailing slash is the `/` character at the end of a URL path (`/about/` vs `/about`). To Google, the two forms are different URLs and can serve different content. Pick one canonical form site-wide and 301 the other. Mixing both produces duplicate-content signals and splits internal link equity.

• TrustRank

  TrustRank is a 2004 Yahoo Research algorithm for separating reputable sites from spam, based on graph distance from a manually curated seed set of trusted sites. Never an official Google ranking factor, but its core idea — trust propagates through links — informs Google's trust signals.

U

• UGC link (UGC)

  A UGC link is any outbound link inside user-generated content marked with `rel="ugc"`. Google introduced the attribute in September 2019 to let publishers signal that a link came from a comment, forum post, review, or similar user-submitted area. Treated as a hint, similar to nofollow.

• URL Rating (UR)

  URL Rating (UR) is Ahrefs's per-page authority score, on a logarithmic 0-100 scale. It's the page-level analog to Domain Rating, computed from the quantity and quality of internal and external backlinks pointing at a specific URL. Useful for comparing individual pages, not whole sites.

• URL slug (URL)

  A URL slug is the human-readable end portion of a URL path — the part after the last slash that identifies the specific page. Best practice: short, lowercase, hyphen-separated words that describe the page and include the primary keyword when natural. Stable slugs preserve link equity over time.

• UTM parameters (UTM)

  UTM parameters are five query-string tags (utm_source, utm_medium, utm_campaign, utm_term, utm_content) that GA4 reads to assign acquisition source, medium, and campaign. They are the source-of-truth for non-organic channels — without them, paid and email traffic frequently lands in Direct.

V

• Vector search

  Vector search is retrieval by similarity in a high-dimensional embedding space, typically using cosine similarity or dot product. It returns semantically related documents even when no keywords match — the retrieval layer beneath most RAG pipelines and modern semantic search products.

• Video carousel

  A video carousel is a SERP block of video results, usually 3-6 thumbnails in a horizontal row, shown for queries with video intent. YouTube dominates algorithmically — non-YouTube videos compete only when properly marked up with VideoObject schema and listed in a video sitemap.

• Video schema

  Video schema is the schema.org/VideoObject structured-data type. It tells Google what a video is, when it was uploaded, how long it runs, what thumbnail to show, and where the playable file lives. Required for video rich-result eligibility, key-moments, and live-stream badges.

• Video SEO

  Video SEO is the practice of making video content discoverable in Google's video results, the Videos tab, and on-SERP video carousels. It covers host choice (self-host vs YouTube vs both), VideoObject schema, video sitemaps, transcripts, and Google's key-moments feature.

• Video sitemap

  A video sitemap is an XML sitemap extension that lists each page on the site that contains a video, along with the video's metadata: thumbnail, title, description, duration, content URL, player URL, age restriction, and tags. It's the single cheapest way to surface video content to Google.

• Viewport meta tag

  The viewport meta tag is the single HTML element that tells mobile browsers how to scale and lay out a page: `<meta name="viewport" content="width=device-width, initial-scale=1">`. Without it, the page renders at a desktop default width (~980px) and gets flagged as not mobile-friendly under mobile-first indexing.

W

• Web Rendering Service (WRS)

  The Web Rendering Service (WRS) is Google's rendering subsystem, an evergreen Chromium-based engine that executes JavaScript on pages Googlebot fetches. It's the "second wave" of indexing — pages with JS-dependent content are first parsed, then queued for WRS to render before final indexing.

• White-hat SEO

  White-hat SEO is search optimization aligned with Google's Webmaster Guidelines. Tactics that improve content quality, technical health, and earned authority. Lower individual leverage than black-hat shortcuts but compounding, low-risk, and the only sustainable approach for real businesses.

X

• X-Robots-Tag

  X-Robots-Tag is an HTTP response header that delivers crawler directives at the server level. Functionally equivalent to a meta robots tag but works on any resource, including PDFs, images, and other non-HTML files where meta tags cannot be embedded. Supports the same directives: noindex, nofollow, etc.

• XML sitemap

  An XML sitemap is a machine-readable file listing URLs that a site wants search engines to discover. A single sitemap holds up to 50,000 URLs or 50 MB uncompressed; larger sites use a sitemap index that references multiple sitemap files.

Y

• YMYL (YMYL)

  YMYL stands for "Your Money or Your Life" — the class of queries where low-quality content can hurt users' health, finances, safety, legal status, or civic participation. Defined in Google's Search Quality Rater Guidelines. YMYL pages are held to a higher E-E-A-T standard than ordinary content.

• YouTube SEO

  YouTube SEO is the practice of optimizing videos for YouTube's own search and recommendation systems — distinct from Google video Search. It covers title, description, tags, chapters, custom thumbnails, end screens, and the watch-time and engagement signals that drive YouTube's algorithms.

Z

• Zero-click search

  A zero-click search is a SERP query where the user gets their answer without clicking through to any result. Already the majority of all Google searches before AI features (SparkToro, 2022: ~65% of US searches). Climbing further with AI Overviews and on-SERP answer surfaces.