Every term we use with clients, explained in plain English, from crawl budget to AI Overviews. No jargon for the sake of it.
Content built to beat everything currently ranking for a keyword, not match it. The "10 times" figure isn't meant literally. It's a quality bar, and the point of it is that a page which only equals the existing results has given Google no reason to promote it.
Tells browsers and search engines that a page has moved permanently, and carries most of its ranking signals across to the new URL. This is what stops a migration from wiping out traffic when pages are retired, merged or renamed.
Signals that a page has moved for now but will return, so search engines generally keep indexing the original URL rather than passing its ranking signals on. Using a 302 for a permanent move is a common technical SEO mistake, and it quietly costs rankings during migrations.
Sent by a server when a page hasn't changed since it was last crawled or requested, letting the browser or bot reuse its cached copy rather than download it again, which speeds up crawling and loading on large, frequently crawled sites.
The "page not found" status, returned when a URL doesn't exist. A few are normal on any site. Large numbers usually mean broken internal links or a botched migration, and they burn crawl budget while sending users into dead ends.
Confirms a page was deliberately and permanently removed, rather than simply missing, which is what a 404 says. It's a stronger and faster signal for search engines to drop the URL from the index. Use it for discontinued products and pages you'll never bring back or redirect.
Running two versions of a page, headline or call to action side by side to see which converts better. It settles arguments with data instead of opinion, which is most of the value.
A stripped-back HTML framework from Google for building near-instant mobile pages. It used to be a requirement for Google's Top Stories carousel, but that was dropped in 2021 when Core Web Vitals became the mobile speed benchmark. AMP is optional now, not a ranking necessity.
The number of times a paid ad is displayed, whether or not anyone clicks it, which makes it a paid media metric rather than an organic one.
Making a website usable by disabled people under the US Americans with Disabilities Act. The UK equivalent obligation sits under the Equality Act 2010, usually benchmarked against WCAG standards. Past the legal case, accessible sites tend to have cleaner code and better usability, which helps SEO.
Optimising a brand for AI Overviews, large language models and platforms like ChatGPT and Perplexity, alongside Google rankings, so it gets found, cited and recommended where buyers now search. Others use the acronym for "Answer Engine Optimisation", a close cousin focused on direct AI answers, so check which a source means.
A bot that crawls websites to gather training data or live information for an AI model or AI search feature. OpenAI's GPTBot, Anthropic's ClaudeBot and PerplexityBot are examples. Blocking them in robots.txt keeps a brand out of AI answers entirely, so allow or block is now a strategic decision rather than a technical default.
Google's AI-generated summary at the top of many results, pulling several sources into a direct answer that sits above the organic listings. Appearing inside one, or being cited as a source for it, is becoming as valuable as a top-three ranking, sometimes more so.
Search behaviour happening inside AI platforms and AI-powered features, ChatGPT, Perplexity, Gemini and Google's AI Overviews, rather than on a traditional results page, and now a channel in its own right alongside Google Search for buyers researching and choosing brands.
A short written description added to an image's HTML so screen readers can describe it to visually impaired users and search engines know what it shows, which also helps images rank in Google Image search and supports accessibility more broadly.
The clickable, usually underlined words in a hyperlink. Search engines read it as a clue about what the linked page covers, so a backlink profile with varied, natural anchors looks organic and helps pages rank for the right terms, where the same exact-match keyword every time does not.
Search tools built to give one synthesised answer to a query rather than a list of links. Perplexity and Google's AI Overviews are the obvious examples. They're changing what ranking means, because the prize is a citation rather than a click.
An old black hat tactic: rewriting existing articles, by hand or with basic software, to pass them off as new and dodge duplicate-content detection. Algorithms got much better at spotting thin, reworded text, so spun articles rarely rank and put trust at risk. Original writing is essential.
Republishing an article on other websites to reach a wider audience, common in digital PR and content distribution. With a canonical tag or clear attribution back to the original, it extends reach and avoids duplicate-content problems. Done carelessly, it dilutes rankings for the source page.
Text produced automatically by software or AI with little or no human input or editing, usually at scale. Google's spam policies specifically target auto-generated content made to manipulate rankings rather than help users, so AI tools need human expertise applied on top before anything goes live.
Links from other websites pointing to yours, and still one of Google's clearest trust signals: a vote of confidence from another site. Earning relevant, high-quality ones through real digital PR and outreach, rather than buying or spamming them, is core to sustainable ranking growth.
Microsoft's free platform for monitoring how Bing crawls and indexes your site, covering indexing status, crawl errors, backlinks and keyword performance. It's Bing's equivalent of Google Search Console, and since Bing also powers some AI search results, check it alongside Google's data rather than instead of it.
Microsoft's crawler for Bing, discovering and indexing pages across the web much as Googlebot does, following links, reading robots.txt and rendering pages, so technical SEO fixes made for Google generally help it too.
Tactics that try to manipulate rankings in ways that break search engines' guidelines: keyword stuffing, cloaking, hidden text, bought links. Gains tend to be short-lived, and the risk of a manual penalty or algorithmic demotion is real. Sustainable rankings come from legitimate methods instead.
The historical measure of visits where someone left a page without interacting further. GA4 replaced it with engagement rate, which credits meaningful time on page, scrolling or extra page views. That's a more useful signal, because a high old-style bounce rate was never automatically bad. A blog reader might have got exactly what they needed.
Content made to build a brand's reputation, awareness or personality rather than to sell a specific product directly, covering thought leadership articles, video series and digital PR campaigns, and playing the long game that builds the recognition and trust behind branded searches.
Search terms containing a company or product name, such as "Patagonia returns policy" or "Nike trainers". High branded search volume is a strong sign of growing brand awareness, and ranking well for your own name, including against competitors bidding on it, matters for reputation and click-through.
A trail near the top of a page showing where the user sits in the site structure, for example Home > Blog > SEO Glossary. It helps people orient themselves, gives search engines a clean read on your hierarchy, and can show up as a rich snippet in the results.
Built purely to funnel visitors or link equity through to another page while adding almost nothing itself, a black hat tactic from old affiliate and PPC schemes that modern search algorithms are good at identifying and demoting.
Any hyperlink pointing at a page that no longer exists, usually returning a 404. Broken links harm user experience and waste crawl budget, so regular audits to fix or redirect them are good technical housekeeping. Finding other sites' broken links is a link-building tactic in its own right.
Stored snapshot of a webpage that a search engine or browser keeps so it can serve it quickly without fetching the live version. Google retired its public "cache:" search operator in 2024, but caching still runs behind the scenes to speed up crawling and page delivery.
An HTML tag, rel="canonical", telling search engines which version of a URL is the main one when duplicate or very similar content sits at several addresses, so ranking signals consolidate on the preferred version instead of being diluted across copies.
Whichever version of a page search engines choose to index and rank when duplicate or near-identical URLs exist, say with and without "www", or with tracking parameters attached. You set it with a canonical tag. Get it wrong and ranking signals split across several URLs.
Showing search engines different content from what real visitors see in order to manipulate rankings, a clear black hat violation of Google's guidelines that can bring a manual penalty or complete de-indexing, well beyond any short-term gain.
Two brands or websites mentioned together in the same piece of content without necessarily linking to each other. Search engines can treat these unlinked mentions as a weaker trust signal, associating your brand with others in its space. Part of why digital PR coverage has value even without a link.
Certain words and phrases habitually appear near each other across the web, "SEO" sitting close to "rankings" or "organic traffic", and search engines read those patterns as a signal of topical relevance, part of the reason thorough content outperforms keyword-stuffed pages.
Another name for auto-generated content, text produced by software or AI with minimal human oversight and usually published at scale rather than to help readers, which Google's spam policies treat as a violation when it exists mainly to manipulate rankings.
Servers spread across multiple locations that deliver website content from whichever one is closest to the visitor. Load times drop, which matters for user experience and Core Web Vitals. A CDN also absorbs traffic spikes, so the site stays reliable when it gets busy.
Comparing your content and rankings against competitors to find topics, keywords and intents they cover and you don't. The output is a prioritised list of what to write next, based on visibility you are actually missing rather than a brainstorm.
A structured cluster of pages around one core topic: a central pillar page linking out to related subtopic pages, which link back. Hubs help search engines read topical authority and depth, and let users explore everything you know about a subject from one place.
How closely a piece of content matches what a searcher actually wants when they type a query, meaning the right subtopics covered, the real question answered and search intent met rather than the target keyword simply being present.
The back-and-forth, natural-language way people search in AI chat tools and voice assistants, where each follow-up builds on the last, much like a real conversation. It rewards content that answers questions directly and in full, over content built around short keyword phrases.
Improving a site so more of the visitors it already has enquire, buy or sign up, instead of chasing extra traffic. It follows on from SEO and AI Search: ranking first counts for little if the people who arrive never convert.
Google's page experience metrics: loading speed (Largest Contentful Paint), responsiveness (Interaction to Next Paint) and visual stability (Cumulative Layout Shift). They're a confirmed ranking factor, though the bigger point is that they track how frustrating or smooth a site actually feels to use.
Your most important, comprehensive pages on the core topics, the ones you'd want a new visitor or a search engine to find first. Close cousin of a content hub's pillar page. These get first call on internal links, updates and promotion because they do the most commercial heavy lifting.
How many pages a search engine will crawl on your site in a given period, set by its size, health and authority. Mostly a large-site problem. On a site with thousands of URLs, budget burned on low-value pages delays new and updated content getting indexed.
How easily a crawler can reach and move through a site's pages, governed by site structure, internal linking, robots.txt rules and server errors. It's the floor everything else stands on, because a page that can't be crawled won't normally be indexed or shown in results.
An automated bot, Googlebot being the obvious example, that works its way through the web by following links, reading page content and code, and feeding what it finds back into a search engine's index or a third-party tool's database.
The whole path from first hearing about a brand to buying from it and beyond, usually across several searches, sessions and channels. For SEO it sets the brief: someone reading a top-of-funnel blog post needs different content and a different call to action than someone on a bottom-of-funnel product search.
Earning coverage, backlinks and brand mentions by pitching journalists and publishers a real story, dataset or campaign rather than asking them for a link. It remains one of the most effective and sustainable ways to build the authority that search engines, and now AI platforms, reward.
Links with no rel="nofollow" or "sponsored" attribute on them pass ranking value from one site to the next. Most organic backlinks are dofollow by default, and they're what digital PR campaigns exist to earn, because they signal trust and authority to Google.
A third-party metric, scored 0 to 100, estimating how strong a site's backlink profile is against everything else in the provider's index. Handy for sizing up competitors or prospects. It is not a Google ranking factor, so treat it as a benchmark rather than a target to chase.
How a site's URLs, subdomains and folders are organised so related content sits together, which helps Google crawl and understand the hierarchy, helps people find their way around, and keeps authority flowing to the pages that matter most for rankings.
Low-quality pages built purely to rank for a search term and push visitors somewhere else, often duplicated across dozens of near-identical URLs targeting slight keyword variations. Google names them in its spam policies and penalises them, because they offer users nothing.
Identical or near-identical content sitting on more than one URL, either within a site or across domains. It muddles which version should rank, splits ranking signals and wastes crawl budget. Canonical tags, redirects or merging the pages into one authoritative version will sort it.
How long someone stays on a page after clicking through from search results, before heading back to them. It gets cited a lot as a user-experience signal, but Google has never confirmed it as a direct ranking factor. Better used as an engagement metric than an optimisation target.
Web addresses generated on the fly from database parameters, such as ?id=123&category=shoes, common on ecommerce and CMS-driven sites. Nothing inherently wrong with them for SEO, though long, parameter-heavy URLs can hurt crawl efficiency and clarity. Static, keyword-friendly URLs are usually the cleaner choice.
Google's framework for judging content quality: Experience, Expertise, Authoritativeness and Trustworthiness. It carries the most weight on YMYL topics. Strong signals here, real author expertise, credible citations and a track record you can point to, increasingly decide whether AI platforms trust a brand enough to cite it.
A backlink earned because a journalist, blogger or publisher rated your content or data enough to reference it. Nobody asked, paid or swapped for it. Editorial links are the gold standard in digital PR, and the natural authority they carry is far harder for competitors to replicate.
Content built to flatter or spotlight other people, brands or sites, the "Top 50 Industry Experts" list being the classic, in the hope they share it and link back. A legitimate outreach tactic done well, though it works best attached to useful data rather than flattery alone.
Contacting journalists, bloggers or site owners directly to pitch content, data or a story, hoping for coverage or a backlink. It's the engine room of digital PR, and results come from relevant targeting and a newsworthy angle rather than volume.
Optimising for how search engines understand real-world things, people, places, brands and concepts, as connected entities in a knowledge base rather than as keyword strings, an approach now central to ranking in both traditional search and AI-generated answers.
The first page someone lands on in a session, homepage or not, tracked in analytics to show which content and channels are actually bringing people in and where SEO and conversion attention should go.
Content written to stay accurate and useful long after its publish date, rather than tied to a moment. It keeps earning traffic and links for years on very little upkeep, which makes it the cheapest thing in most content plans.
A link from your page out to another website, as opposed to an internal link that stays on your own. Pointing readers at credible, relevant sources gives them useful context and does your page no harm with Google.
The filter and sort options on ecommerce and listing sites, covering size, colour, price and brand, where each combination often spawns its own URL. Leave it unmanaged and you get thousands of thin, duplicate URLs eating crawl budget. Canonicalisation has to be set up properly.
Sometimes called position zero, the highlighted answer box Google puts above the standard organic results, pulling a short answer as text, a list or a table from a page that already ranks. It earns strong visibility and click-through, and it's the clearest structural link between classic on-page SEO and how AI Overviews are generated.
A report, tool or whitepaper put behind a form that asks for an email address or other details before anyone can read it. Useful for generating leads. The trade-off is that search engines and casual visitors can't see it either, so it earns no organic visibility.
Another name for a doorway page, built solely to rank for a search term and funnel visitors elsewhere with little value of its own, and Google treats it as a spam violation that can earn the site a manual penalty.
Restructuring content so generative, conversational AI platforms can cite a brand outright as the definitive answer to a complex buyer query rather than listing it as one blue link among several, a discipline also called Answer Engine Optimisation.
A free Google tool that emails you when new content matching your chosen keywords appears online. Agencies use it to watch brand mentions, keep track of competitor coverage, and catch unlinked mentions of a client that outreach can turn into a backlink.
Google's vast, constantly evolving system of ranking factors and machine-learning models, which decides which pages best answer a query and in what order they appear. How it weighs relevance, authority and user experience sits underneath every SEO decision.
Google's free web analytics platform, currently GA4, for measuring traffic, user behaviour and conversions. It shows where visitors came from, what they did on site and whether any of it turned into enquiries or sales. Without it you're guessing whether SEO work is having real impact.
The dropdown of suggested searches Google offers as you type, generated from queries real people make, and a free source of long-tail keyword ideas that shows the exact phrasing your content needs to answer.
A largely defunct manipulation tactic: get enough sites linking to a target page with identical keyword-rich anchor text and Google would rank it for that term regardless of relevance. Later algorithm updates put a stop to it.
Free listing that controls how a business appears in Google Maps and local search results: name, address, opening hours, reviews and photos. Optimised and actively managed, it's one of the strongest levers for winning visibility in local "near me" searches.
A 2010 rebuild of Google's indexing infrastructure that replaced periodic batch updates with continuous crawling and indexing. New and updated content started appearing in results far faster. It changed the plumbing, not the ranking criteria.
An early-2000s term for the visible fluctuation in rankings while Google's index updated across its data centres, throwing up unpredictable results for a period, something modern continuous indexing has largely consigned to history.
A major 2013 rewrite that moved Google from matching individual keywords towards understanding the intent and meaning behind a whole query. It laid the groundwork for conversational and semantic search, including the natural-language queries people put to AI-driven search today.
Google's vast database of real-world entities, people, places, organisations and concepts, plus the relationships between them. It powers knowledge panels and direct answers, and being represented clearly in it helps a brand appear accurately across both classic and AI-powered search.
The information box that appears alongside search results, or above them on mobile, summarising facts about a recognised business, person or brand, mostly drawn from the Knowledge Graph. Claiming yours and correcting what it says strengthens brand visibility and trust.
A major 2011 update aimed at thin, low-quality and duplicate content, built to reward well-written, useful pages and demote content farms. Panda was folded into Google's core algorithm in 2016, and it's still a large part of why content quality isn't negotiable in SEO.
Loss of search visibility, applied either by hand by Google's spam team for a guideline breach or algorithmically when an update devalues manipulative tactics. Recovery means identifying the cause and fixing the underlying issue, then submitting a reconsideration request if the penalty was a manual one.
Launched in 2012 to target manipulative link building: spammy, low-quality or keyword-stuffed backlinks built purely to game rankings. It ran as a periodic update at first, then became a real-time part of Google's core algorithm in 2016, so sustainable earned links matter more than ever.
Rolled out in 2014, tying local search results far more closely to core ranking signals, so content quality and backlinks began counting much more towards local pack and Maps rankings than before.
Theorised effect, never officially confirmed, where new websites struggle to rank competitively for a period even with good SEO, until Google has built up enough trust and history in the domain. Deliberate filter or natural lag, nobody can say. Patience and consistency are the fix either way.
Free Google tool showing how a site performs in organic search: impressions, clicks, average position, indexing status and technical issues Google has flagged. A direct line to its own data, and the basis of most technical SEO analysis.
Also known as the Page Layout Algorithm. This 2012 update demoted pages that crammed so many ads above the fold that the real content sat below the visible screen on load, and it pushed sites towards user-friendly layouts that put content ahead of advertising.
Google's official rules on what counts as acceptable practice for getting a site crawled, indexed and ranked, covering technical, content and spam standards. Now rebranded as "Google Search Essentials", they remain the closest thing to a rulebook for avoiding penalties and staying in Google's good graces.
The name Google Search Console went by before its 2015 rebrand. Functionally the same free platform for monitoring indexing, search performance and technical issues, so the old name in older guides or forums means Search Console.
Google's automated crawler, which discovers, fetches and analyses pages so they can be added to the index and considered for ranking. How efficiently it can work through a site depends on speed, internal linking and technical health, and that decides how quickly new content gets found.
Tactics that sit between fully compliant white hat practice and clearly manipulative black hat work, bending Google's guidelines without quite breaking them. The risk is real, because algorithms and enforcement keep closing these grey areas, and yesterday's grey hat tactic is often tomorrow's penalty.
Writing and publishing content on someone else's website, usually in exchange for a backlink and exposure to a new audience. On relevant, reputable sites, with content worth reading, it stays a legitimate digital PR tactic. Done at scale purely for links, it's a spam risk.
Link-building tactic that offers a relevant site a data-led infographic plus a pitched written introduction as ready-to-publish guest content, pairing visual appeal with tailored outreach to improve the odds of a real placement and backlink.
Main heading tag on a page, normally the visible page title. It tells readers and search engines what the page is about, so use the primary keyword naturally and keep it to one H1 per page for a clear content hierarchy.
The H1 to H6 tags that structure a page from main heading down to subheadings, helping readers scan the content and helping search engines work out how its topics relate.
Early 2000s Google concept that picked out "expert" pages, the ones linked to by many topically relevant sites, then used their outbound links to help rank other pages. It's regarded as a forerunner of how authority and relevance in links get weighed today.
Treating SEO as one joined-up discipline rather than a set of isolated tactics, with technical health, content quality, user experience, authority and brand signals all handled together. That matches how modern search engines assess a site overall instead of scoring individual ranking factors in isolation.
HTML attribute telling search engines which language and regional version of a page to show, used on sites publishing for multiple countries or languages, and when implemented correctly it prevents duplicate content issues and serves searchers the right version.
Server response confirming a webpage loaded successfully with no errors. It's what you want crawlers and visitors to get back, since codes like 404 (not found) or 500 (server error) flag problems that hurt crawling and users.
Secure version of HTTP, encrypting data between a browser and a website with an SSL/TLS certificate. Google has confirmed it as a ranking signal and it's now a baseline trust requirement, since browsers mark sites without it "Not Secure", costing credibility and conversions.
Link from another website pointing at yours, also called a backlink. Search engines treat them as votes of confidence, so earning them from relevant, reputable sites builds authority and improves rankings.
When a search engine has indexed far more pages from a site than are useful or unique, usually thin, low-value or duplicate ones. It dilutes crawl budget and drags on site-wide quality signals, making life harder for the pages that actually matter.
Whether a search engine can crawl a page, understand it and add it to its index so it can appear in results. Pages get excluded by accident all the time through stray noindex tags, robots.txt blocks or crawl errors, so this is a fundamental technical health check.
Search where someone wants to learn something rather than buy or reach a specific site, such as "what is domain authority". Reading intent this way shapes the content a site needs, both to rank and to steer people towards a commercial decision later.
Link from one page to another on the same website, used to help search engines discover pages and understand how they relate, spread authority across the site, and guide readers towards related content or a conversion point.
Pop-up or overlay that covers the main content, often appearing on mobile before the visitor can reach the page at all. Intrusive ones hurt rankings as well as the experience, and Google specifically penalises pages where disruptive mobile pop-ups block the content.
Making JavaScript-rendered websites and web apps properly crawlable and indexable by search engines. Google has to render the JavaScript before it can read the content, so sites built on JS frameworks need extra technical care to stop pages being missed or misread.
Two or more pages on a site chasing the same keyword or search intent and competing with each other rather than one strong page winning. Search engines get mixed signals about which to rank, and performance for that term usually weakens as a result.
Grouping related keywords by shared search intent so a single comprehensive page can target the whole group instead of the terms being spread thinly across many pages, which keeps a content strategy efficient, covers topics fully and avoids cannibalisation.
The percentage of a page's word count taken up by a given keyword. Largely outdated, since search engines weigh natural language, relevance and context rather than any particular keyword percentage.
Score from SEO tools estimating how hard page one would be for a given keyword, based on the strength of the pages already ranking there. Useful for separating realistic targets from long-term ambitions.
A website's current position in search results for a specific keyword. Tracking rankings over time shows whether the SEO work is paying off, but treat it as one signal among several rather than the whole scoreboard, since rankings vary by user and location.
How search engines recognise different grammatical forms of the same root word, such as "run", "running" and "runner", as related, which spares content from awkward exact-match repetition.
Overloading a page with a keyword or phrase in an unnatural way to try to manipulate rankings, which is both outdated and against Google's guidelines. Search engines penalise it because it wrecks readability and gives users nothing of value. Firmly a legacy black-hat tactic.
The words and phrases people type into a search engine when they want information, products or services. Everything else in SEO follows from them: what you write, how you structure it, and which audience ends up finding the site.
Standalone page built for visitors to arrive on with a specific goal in mind, whether that's a purchase, a sign-up or a quote request. In SEO it has to satisfy search intent and convert, balancing ranking potential against a clear path to action.
A mathematical technique for finding relationships between words and concepts by analysing patterns across large volumes of text. It comes up constantly in SEO discussions about topical relevance. Its actual role in modern search algorithms is debated.
Content built deliberately to attract attention and earn links naturally, such as original research, data studies or a useful free tool. Legitimate and effective despite the name, which only describes content made to be worth linking to.
The ongoing work of acquiring backlinks from other websites to build a site's authority and its ability to rank. Done properly it now means earning links through real relationships, PR and quality content, rather than manipulative shortcuts that risk a penalty.
The value or authority passed from one page to another through a link, shaped by the linking site's authority, its relevance and how many other links share that page, and the core reason link quality counts for more than link quantity.
Two websites agreeing to link to each other purely to manipulate rankings, rather than because either link adds anything for a reader. Excessive reciprocal exchanges breach Google's guidelines and can trigger a manual action once the pattern starts to look deliberate.
A network of sites built purely to link to each other, or to one target site, with no real content behind them. Google's spam guidelines cover this explicitly. The links carry no ranking value and can bring a penalty with them.
Informal name for the ranking value or authority a link passes from one page to another. Same underlying idea as link equity, just described more casually. Stronger, more relevant linking pages tend to pass more of it.
How many other websites link to a page or domain, once treated as a strong ranking signal in raw numbers but now heavily discounted by search engines in favour of the quality and relevance of the linking sites.
Every backlink pointing at a website, taken together with the quality of the sources, their relevance, the anchor text and how the links have grown over time. A natural-looking profile signals trust. One stuffed with spammy or manipulative links will hold a site back.
Finding broken, lost or unlinked mentions of a brand and turning them into working backlinks. The authority has already been earned, it just isn't being counted, so this is an efficient recovery job. Most sites overlook it in favour of building new links.
Links across the web gradually breaking as pages are moved, renamed or deleted without redirects, affecting both a site's outbound links and the backlinks pointing at it, so regular link audits protect user experience and link equity.
Link building done primarily to manipulate rankings rather than earn links on merit. Google's Webmaster Guidelines define and prohibit it by name, with bought links, excessive link exchanges and link farms as the stock examples. Sites caught risk a manual penalty.
Low-quality, manipulative or irrelevant links built purely to game rankings rather than serve users. Google goes after it with both algorithms and manual review, and sites associated with it can lose rankings or be penalised outright.
The clickable, visible words in a hyperlink, also called anchor text, which give users and search engines context about the linked page, so relevant, natural wording is more valuable than generic phrases like "click here".
The rate at which a site gains or loses backlinks over a given period. Gradual, steady growth looks organic. A sudden spike with nothing behind it raises flags and invites closer scrutiny.
The AI model behind ChatGPT, Perplexity, Gemini and Claude, trained on enormous volumes of text to produce human-like answers to questions. Buyers increasingly put their question to an LLM before they ever open Google, so being described accurately and cited inside those answers now matters as much as ranking on page one.
Another label for optimising content so large language models learn about a brand, understand it and recommend it accurately, covering the same ground as AEO and GEO.
Structured data markup that spells out a business's name, address, phone number, opening hours and reviews for search engines, helping them display accurate, rich information in local results and Google Business Profile listings.
Any mention of a business's name, address and phone number (NAP) on another website, directory or platform, link or no link. Consistency and accuracy across them is what helps search engines trust that the business is legitimate and located where it says.
Map-based block of typically three business listings that appears near the top of Google's results for location-based searches such as "plumber near me". It sits above most organic results and takes a large share of the clicks.
Promoting a business to people searching within a specific geographic area. Broader than local SEO alone, it pulls in reviews, citations and Google Business Profile management so firms with a physical location or a service area get found by nearby customers.
Optimising a website and wider online presence to rank for searches tied to a place, such as "SEO agency Sheffield", through Google Business Profile optimisation, local citations, reviews and location-specific content that brings in nearby customers.
Reading a site's server logs to see exactly how search engine crawlers behave on it, which pages they visit, how often and where they hit errors, revealing crawl budget problems that other tools can't show.
Longer, more specific search phrases, typically lower in volume but higher in intent and less competitive than broad head terms, and often the faster, more cost-effective route to relevant traffic that's already close to converting.
Popular but technically inaccurate name for related words and phrases that add topical context to content. The term borrows from Latent Semantic Analysis, though most practitioners now accept that modern search engines use much more advanced language understanding than true LSA.
A penalty applied by a human reviewer at Google, not an algorithm, when a site breaches the webmaster guidelines through unnatural links or thin content. It appears in Google Search Console with the reason and next steps. Rankings and indexing usually stay suppressed until it's fixed and a reconsideration request is approved.
HTML tag summarising a page in roughly 150 to 160 characters, often used as the snippet beneath the title in Google's results. It isn't a direct ranking factor, but a clear, compelling one lifts click-through rate, so write it yourself rather than leaving Google to pick a sentence off the page.
Legacy HTML tag that once listed the keywords a page targeted, confirmed by Google back in 2009 to carry no weight whatsoever in rankings and used by no major search engine today, so ignore it entirely rather than spending time filling it in.
Also called a meta refresh: an HTML tag that automatically sends visitors and crawlers to a different URL after a short delay. Google advises against it for permanent moves, since it's slower and less reliable at passing ranking signals than a proper server-side 301.
Sits in a page's head section and tells crawlers how to treat that specific page: noindex to keep it out of results, nofollow to stop its links passing value. Page-level control over indexing, without touching robots.txt.
Snippets of HTML in a page's head section that describe its content to browsers and search engines, covering titles, descriptions, robots directives and more. The title tag directly influences rankings. The others mostly govern how a page appears in results or how it gets crawled and indexed.
An exact copy of a website hosted at a different URL, used historically to spread server load, hold backups or serve regions where the original is blocked. Search engines read it as duplicate content unless canonical tags make clear which version should rank.
Google uses the mobile version of a site as the primary basis for crawling, indexing and ranking, reflecting that most searches now happen on mobile devices. Since 2023 this applies to virtually every site, so a poor or incomplete mobile experience directly limits visibility.
Name, Address and Phone number. These core business details need to appear identically across the website, the Google Business Profile and every directory listing, because that consistency is how Google confirms a business is legitimate and correctly located for local search.
Branch of AI that works out the meaning and intent behind human language instead of matching keywords. Google applies it through systems like BERT to read queries more like a person would, and it's the same machinery behind how AI search tools and chatbots understand and answer questions.
A backlink earned organically because another site found your content valuable enough to reference, with no payment, exchange or manipulation involved. Exactly what Google's guidelines ask for, and the foundation of any sustainable digital PR or link building work.
Someone already knows the website or brand they want and uses Google as the quickest way to get there, searching a company name instead of typing its URL, and recognising query types like this helps shape keyword and content strategy.
Malicious attempts to damage a competitor's rankings, whether by building spammy backlinks to their site, scraping and republishing their content elsewhere, or trying to hack it. Google provides defences like the disavow file. Confirmed cases are rarer than site owners tend to fear.
Link attribute (rel="nofollow") telling search engines not to pass ranking credit through that particular link. It was introduced to combat comment spam and now sits alongside sponsored and ugc, all of which Google treats as a hint rather than an absolute rule.
Meta robots directive, or HTTP header, that keeps a specific page out of the search index. The page can still be crawled and its links still followed, it just won't appear in results. Standard on thank-you pages, filters and duplicate content.
Value added to a link's rel attribute (rel="noopener") when it opens in a new tab, preventing that new page from using JavaScript to access or control the original tab, a browser security best practice rather than anything that affects rankings.
Rel attribute value that stops the browser passing referrer information to the page a link points to. It also implies noopener, so the destination cannot reach back into the tab that opened it. Used for privacy and security. No direct ranking effect.
Label GA showed in place of the actual search term once Google began encrypting search data in 2011, near-universal after the shift to HTTPS by 2013. Organic visits were still visible, the keyword that drove them wasn't. Google Search Console now helps fill that gap.
Everything done away from your own website to build its authority and reputation in search engines' eyes: backlinks first and foremost, plus brand mentions, digital PR and social signals. One of the three core pillars of organic visibility, alongside on-page and technical SEO.
Optimising the content and HTML of individual pages, from titles and headings to body copy, internal links and images, so both users and search engines can tell what a page is about and find it useful. Sits with technical and off-page SEO as one of the three foundational disciplines.
Meta tags originally created by Facebook that control how a page's title, description and image appear when it's shared on social platforms. They don't influence Google rankings directly, but they decide whether a shared link looks professional and clickable, supporting social visibility and traffic.
The unpaid listings on a search engine results page, shown because they're relevant to the query and ranked by the algorithm rather than by advertising spend, distinct from the paid ads alongside them and the traffic source SEO exists to grow.
Visits arriving from unpaid search listings, as opposed to paid ads, social, referral or direct traffic. It's the headline metric SEO activity aims to grow, though raw visit counts matter far less than the leads and revenue sitting behind them.
No internal links point to it, so search engines and users have no route to it through normal browsing or crawling. Orphan pages go unindexed or underperform because nothing signals that they exist.
Points from your page out to a different domain. Linking to relevant, authoritative sources supports content credibility and user experience, and is a normal part of well-written content. Backlinks run the other way, into your site.
How quickly a page's content loads and becomes usable, measured through metrics such as Core Web Vitals. Google has confirmed it as a ranking factor, if a modest one, and it drives user experience and conversion rate. Slow pages lose visitors before anything is even seen.
Google's original algorithm for scoring a page's importance by the quantity and quality of links pointing at it, built by founders Larry Page and Sergey Brin. The public PageRank toolbar was retired in 2016, but link equity still shapes how the algorithms work today.
Backlink obtained in exchange for money, products, services or other compensation, with no nofollow or sponsored tag disclosing the arrangement. Google's guidelines treat undisclosed paid links as manipulative, so the likely outcome is a manual action rather than lasting ranking gains.
An expandable Google SERP feature that stacks related questions, each revealing a short answer sourced from a webpage when clicked. Appearing there lifts visibility and click-through, and the questions themselves are good raw material for content and FAQs.
Covers a broad topic in depth and links out to a cluster of narrower related pages, which link back to it. That hub and spoke structure helps search engines read topical relevance and lets users work through a subject from one place.
A searcher clicks a result, bounces straight back to the results page and picks a different one instead. Some believe the pattern signals to Google that the first result didn't satisfy the query. Google has played it down as a direct ranking factor, but it still flags a poor content-query match.
The main search term a page is optimised to rank for, usually reflected in its title tag, H1 and URL, with supporting or secondary keywords woven in around it to broaden relevance without pulling focus off the main target.
Group of sites, often built on expired domains that still carry authority, created purely to link back to a "money site" and manipulate its rankings. Google actively targets PBNs as a guidelines violation, so the usual outcome is a manual action and lost rankings.
The question or instruction a user types or speaks into an AI tool like ChatGPT or Perplexity to get an answer. Knowing the real prompts buyers use to research a purchase, rather than the keywords they'd type into Google, is central to appearing in AI-generated answers.
Google temporarily favouring newer content for topics where recency clearly matters, such as breaking news, trending events or annual updates, even when older pages are otherwise more authoritative, an effect that fades once the topic settles and intent shifts back to depth.
The technique many AI search tools and chatbots use to pull real-time information from the web before generating an answer, instead of relying only on their training data, then cite the sources used. It's the mechanism that makes a website's content eligible to be quoted inside an AI answer.
Machine learning inside Google's core algorithm, launched in 2015, which interprets unusual or never-seen-before queries by matching them to concepts Google already understands. One of Google's earliest steps toward AI-driven search, and the groundwork for the AI-powered ranking systems used today.
Two websites agreeing to link to each other. Done naturally it's harmless, but building links purely through swaps violates Google's guidelines and looks manipulative at scale. Real authority comes from earning links organically, not trading them.
A formal appeal submitted through Google Search Console after fixing the issues behind a manual action penalty, asking Google to review the site and lift the restriction. Typically the final step in recovering visibility lost to a guideline violation.
Alternative queries Google shows near the bottom of a results page, generated from what other users searched alongside the original term, and a handy free source of keyword ideas that shows how searchers phrase similar needs.
Shortened web address that omits the domain and gives only the path from the current page, /services/seo rather than the full https://example.com/services/seo. Used in internal linking, though absolute URLs are generally safer for canonical tags and technical SEO.
Curated lists of helpful links on a specific topic that other websites publish for their audience, and a real link building opportunity: if your content deserves a place, outreach to relevant resource pages can earn a natural, relevant backlink.
Standard organic search result enhanced with extra visual detail pulled from structured data on the page, such as star ratings, prices or cooking times. Rich snippets can lift click-through rate sharply, making them a technical SEO win that has nothing to do with ranking position.
A plain text file in a website's root directory telling search engine crawlers which pages, folders or resources they're allowed and disallowed to crawl. Foundational technical control, and easy to get wrong: a bad rule can accidentally block important pages from being indexed.
Standardised code vocabulary added to a page's HTML that spells out exactly what its content means to search engines, whether that's a review, a recipe, an event or a product price. It underpins rich snippets and many SERP features, and sits at the core of technical SEO work.
The set of rules and calculations a search engine uses to decide which pages best answer a query and in what order, weighing hundreds of signals in Google's case, from relevance and authority to page experience, and updated constantly.
Black-hat technique where attackers manipulate search results, often by exploiting trending topics or hacking legitimate sites, to trick people into clicking links that lead to malware, scams or phishing pages rather than real content.
Whatever a search engine displays after someone runs a query: the mix of organic listings, ads, and features like maps or featured snippets. Ranking higher on the SERP for relevant terms is the fundamental goal of SEO.
The underlying reason behind a query, whether someone wants information, a specific website, a comparison of options, or to buy. Google prioritises content that matches intent over content that simply repeats keywords, so intent decides what you create and how you structure it.
Individual organic listings returned for a query, each showing a page title, URL and description snippet. They sit inside the wider SERP alongside paid ads and other features, and making them more relevant and appealing is how clicks are earned.
Exact word or phrase someone types into a search engine, as recorded in tools like Google Search Console. Distinct from a "keyword", which usually means the topic a business targets. Search terms are the real, sometimes messy, language people actually use.
A metric estimating what share of the possible clicks or impressions a website captures across all the keywords it ranks for, expressed as a percentage. Useful for tracking organic performance over time, as long as it's read as one signal among many rather than the whole story.
Average number of times a keyword is searched over a given period, usually monthly, within a chosen location. It helps prioritise which keywords are worth targeting, though volume alone guarantees nothing. Competition and intent matter just as much.
Supporting search terms closely related to a page's main target keyword, woven naturally through the content to capture additional relevant searches and show search engines the full scope of the page.
The security protocol that encrypts data passing between a browser and a website's server, shown by the padlock icon and an https:// address. Largely succeeded by TLS, though everyone still says SSL. Having it is both a trust signal and a confirmed, if minor, Google ranking factor.
Small set of broad, obvious terms describing a business or topic, used to kick off keyword research before tools and researchers expand them into a far wider list of ideas.
Search that interprets the meaning and context behind a query rather than matching the literal words typed, using natural language understanding to work out what a searcher actually wants. Modern Google results and AI-generated answers are both built on it.
Search Engine Optimisation: improving a website so it ranks higher in organic, that is unpaid, search results. It combines technical fixes, quality content and authority building to help the right people find a business first, driving sustainable traffic without paying per click.
A thorough review of a website's technical health, content quality and backlink profile, used to uncover what's holding rankings back and where the biggest opportunities lie, and typically the first step in building an SEO strategy.
Organising a website so content on the same topic is grouped together and interlinked, with a broad "pillar" page connecting out to more specific supporting pages. It signals topical authority to search engines and makes related content easy to move through.
Extra elements search engines display alongside standard organic listings: featured snippets, People Also Ask boxes, image packs, local packs and more. They can push organic listings further down the page, so optimising to appear within them is now part of search visibility.
Emerging metric estimating how often a brand is mentioned, cited or recommended across AI platforms' answers for a given topic, compared with its competitors. Share of voice for the AI era. As more buyer research happens inside AI chat interfaces, tracking it matters as much as tracking rankings.
Percentage of the total available search visibility, clicks or impressions a brand captures for a defined set of keywords, measured against its competitors, which makes it a benchmark for market share in search rather than a number read in isolation.
Broad search terms of one or two words, "SEO agency" being the obvious example. High volume, high competition, and vague intent behind them. They are harder to rank for and harder to convert than more specific long-tail alternatives.
The extra links Google sometimes shows beneath a site's main search result, usually on branded or navigational searches, pointing to pages like "About" or "Contact". Google's algorithm generates them automatically. You cannot request them or buy them.
A file, usually XML, listing the URLs on a site that matter, so search engines can discover and crawl them and understand the structure more efficiently. Submitting one through Google Search Console is a small technical job with real payoff on larger or newer sites.
A backlink that appears on every page of a site, usually in a footer, sidebar or template, the "website by" credit being the classic case. Search engines can flag these as low-value or manipulative when they exist to pass SEO benefit rather than because anyone needs the link to find their way around.
People increasingly look for information, reviews and recommendations inside social platforms like TikTok, Instagram and YouTube instead of starting on Google. Optimising content and presence for that behaviour is turning into its own discipline, sitting alongside traditional and AI search.
Manipulating search engines through deceptive tactics: keyword stuffing, cloaking, hidden text, artificial link schemes. All of it aims to rank higher than the quality of the content deserves, all of it breaches search engine guidelines, and it risks a manual penalty, so it has no place in an SEO strategy meant to last.
rel="sponsored" is the HTML attribute added to paid, affiliate or advertorial links to tell Google the placement was bought rather than editorially given. Applied properly it keeps a site inside search engine guidelines and cuts the risk of a manual penalty.
An HTML image attribute that hands the browser several versions of an image at different sizes and lets it pick the most appropriate one for the visitor's screen, which speeds up page load and supports strong Core Web Vitals scores.
A standardised way of labelling the content in a page's code, using a shared vocabulary such as schema.org, so search engines can work out exactly what it means rather than only what it says. It is the mechanism behind rich snippets and a good deal of other SERP features.
A prefix placed before the main domain to create a distinct section of a site, such as blog.example.com. Search engines can treat a subdomain as a separate entity from the root domain for crawling and authority, so many sites now keep SEO-critical content in subfolders instead.
Organising a website's content into categories, subcategories and tags that form a logical, crawlable structure, which helps search engines see how pages relate to one another and helps users find what they need without getting frustrated.
The infrastructure side of SEO: crawlability, indexability, site speed, security and structured data. Get this wrong and nothing else lands, because content and links both depend on search engines being able to find, understand and rank what is there.
Term Frequency-Inverse Document Frequency, a formula measuring how important a word is to a document by weighing how often it appears there against how common it is across all documents, used by some SEO tools to suggest terms your content might be missing.
Pages with little real value, depth or originality: sparse product descriptions, duplicate text, auto-generated filler, none of it much help to the person searching. Google's quality systems are built to identify thin content and hold it back in rankings.
A technique that builds layers of links pointing at your existing backlinks rather than directly at your site, on the theory that it boosts their authority. Higher risk and largely outdated now, having fallen out of favour as Google got much better at spotting manipulative link networks.
The HTML element that names a page. Search engines show it as the result headline and browsers put it on the tab. Google rewrites it when it thinks it can do better, but it remains one of the strongest on-page relevance signals you control, so every page needs its own.
The final segment of a domain name, sitting at the top of the internet's naming hierarchy: .com, .org, or a country code like .co.uk. The choice can carry trust signals, and a local TLD such as .co.uk can help reinforce that a site targets the UK market.
How closely a page or a whole site matches a subject, judged on the terms, entities and related concepts it covers rather than keyword matches alone. Covering a topic properly is what reads as expertise to a search engine, and it lifts rankings across the wider set of related queries.
A search made with clear intent to buy or complete an action, along the lines of "buy running shoes online" or "book an SEO audit". Spotting these tells you which pages need optimising to convert visitors rather than only to rank.
The encryption protocol that secures data travelling between a browser and a website, successor to SSL, visible as the padlock icon and "https" in the address bar. Google has confirmed it as a lightweight ranking signal, and it is the baseline for user trust and data safety.
A concept from an early Yahoo and Stanford research paper, describing how trustworthiness might be measured by tracing links back to a set of known, trusted "seed" sites. Not a confirmed live Google ranking factor, though it comes up alongside PageRank whenever people discuss how authority and trust could be modelled.
rel="ugc" marks links inside user-generated content such as blog comments and forum posts, telling Google the site owner did not place them editorially, so search engines can treat them appropriately instead of passing full ranking value.
Google's approach, introduced back in 2007, of blending images, videos, news, maps and shopping directly into standard organic results rather than keeping them in separate silos, which laid the groundwork for the varied SERP features dominating results pages today.
Backlinks created to manipulate rankings rather than earned editorially, meaning paid links, link exchanges and listings on spammy directories, all of which breach Google's Webmaster Guidelines and can trigger a manual action or algorithmic penalty that suppresses rankings once discovered.
A third-party SEO tool metric, scored 0 to 100, estimating the backlink strength of one individual URL from the quantity and quality of links pointing directly at it. Use it to compare the relative authority of specific pages, not to judge a whole domain.
The part of a URL that identifies a specific page, sitting after the domain: "/seo-glossary" in example.com/seo-glossary. Short, descriptive, readable slugs tell users and search engines what a page is about before anyone clicks through.
The reason behind a search: someone wants information, a specific site, a comparison, or they are ready to buy. Working out which one you are dealing with is the difference between content that satisfies a searcher and content that only targets a keyword.
Search specialised to one type of content or one industry, such as Google Images, Google Shopping, YouTube or Skyscanner, rather than the open web, and ranking well inside a vertical usually takes its own dedicated optimisation approach, distinct from standard organic SEO.
Searches spoken aloud to a device or assistant, Siri, Alexa or Google Assistant, instead of typed. They tend to run longer, sound more conversational and arrive as questions more often, which is pushing content optimisation towards natural language and direct, clearly structured answers.
A general term for how trustworthy and influential a website looks to search engines, usually inferred from the strength of its backlink profile. It gets used loosely alongside tool-specific scores such as Domain Rating, each of which measures the underlying idea slightly differently.
A dated web-analytics term for any single request made to a server. Loading a page, an image or a script each counts as one, so a single page view can generate dozens of hits. Sessions and page views have largely replaced it as more meaningful measures.
How a site's pages are organised and interlinked, ideally in a logical hierarchy running from homepage to categories to individual pages, so users can find their way around and search engines can crawl the site and distribute authority efficiently.
Any content or technique used to manipulate rankings in breach of search engine guidelines, from keyword stuffing and cloaking through to link schemes and mass-produced AI junk content. Google runs dedicated systems for detecting and demoting it, SpamBrain among them.
SEO carried out strictly within search engines' guidelines, with useful content, links earned naturally and a technically sound site rather than manipulative shortcuts, slower than black-hat tactics but building sustainable growth that carries no penalty risk.
An HTTP header used as an alternative to the meta robots tag, telling search engines how to handle a file, blocking a PDF or image from being indexed for instance. Because it is applied at server level, it works on file types that cannot carry HTML tags.
Your Money or Your Life pages cover topics like finance, health, safety or legal advice, where inaccurate information could cause real harm to someone's wellbeing or finances. Google holds this content to a higher quality bar, so strong E-E-A-T signals matter more here than almost anywhere else.
A search answered on the results page itself, by a featured snippet, AI Overview or knowledge panel, with no click through to any website. These keep getting more common as AI-generated answers spread, shifting the goal of visibility away from clicks alone towards brand mentions and citations.
