Glossary
85 terms that come up in audits, reports and strategy calls, each explained in a few sentences, with links to the guides that go deeper.
A 301 redirect permanently sends visitors and search engines from one URL to another. Google treats it as a strong signal that the target should be the canonical URL and shows the target in results; for site moves, it recommends a 301 from every old URL to its closest new equivalent, kept for at least a year, rather than blanket redirects to the homepage.
AI crawlers are bots that fetch web pages for AI products, either to show sources in AI search, such as OAI-SearchBot and PerplexityBot, or to train models, such as GPTBot and ClaudeBot. They are separate user agents with separate robots.txt rules, so you can allow search crawlers while blocking training ones; check CDN and firewall bot settings too, and remember that many don’t run JavaScript.
AI Mode is a conversational mode in Google Search for questions that need exploration, reasoning or comparison: it gives an AI-generated answer with links to its sources and takes follow-up questions. It is available in Türkiye in Turkish as “AI Modu”; because it uses query fan-out, a page can be cited for a sub-question the user never typed, often without earning a click.
AI Overviews are AI-generated summaries at the top of some Google results pages, with links to the pages they draw on; in Türkiye they appear in Turkish as “AI Bakışı”. A page needs to be indexed and eligible for a snippet to be shown as a supporting link, with no special markup; Search Console counts these impressions in its normal Web data and in the Search Generative AI performance report.
Anchor text is the visible, clickable text of a link. Search engines use it to understand what the linked page is about, so make it descriptive and concise rather than “click here”; on your own site, point the same anchor text at one page only, and avoid keyword-stuffed anchors in links you place on other sites, which fall under Google’s link spam policy.
Average position in Search Console is the average of the topmost position your site held in results, across all impressions for a query or page. It is easily misread: a page ranking first for a few queries and fortieth for many lands somewhere in between, and all links in an AI Overview share that feature’s single position, so filter by query and page before drawing conclusions.
A backlink is a link from another website to yours. Google uses links to discover pages and as one signal of a page’s relevance and importance, so count relevant referring domains rather than raw links, and remember that links bought to pass ranking credit breach Google’s spam policies.
Bing Webmaster Tools is Microsoft’s free tool for site owners, showing Bing’s view of your indexing, search performance, backlinks and crawl errors. It matters beyond Bing because Microsoft Copilot grounds its answers in Bing’s index; you can import sites from Search Console, and its AI Performance report shows how often your pages are cited in Copilot and Bing’s AI answers.
A branded query is a search that includes your brand or product name, misspellings included, such as “acme pricing”. Branded searches measure demand you already have, so report them separately from non-branded ones, using Search Console’s branded queries filter where your property has it and a regex filter where it doesn’t; otherwise a strong month for brand awareness will look like an SEO win.
A canonical tag (rel=canonical) tells search engines which URL is the main version of content available at several addresses. It is a hint, not a rule: Google may choose a different canonical, especially when your internal links, sitemaps, redirects or hreflang point elsewhere, so keep all of them consistent, and let each language version be its own canonical.
A ccTLD is a country-code top-level domain, such as .uk or .tr, including forms like .co.uk and .com.tr. Google treats it as a strong signal that a site targets that country, but each one is a separate site whose authority has to be built from scratch, so it suits separate legal entities or brands best; a few, such as .io and .tv, are treated as generic.
The Chrome UX Report (CrUX) is Google’s public dataset of how real Chrome users experience websites, including LCP, INP and CLS. It supplies the field data in PageSpeed Insights and Search Console’s Core Web Vitals report, and Google Search uses it for page experience; pages and sites without enough eligible traffic have no CrUX data, so collect your own real-user data as well.
Click-through rate (CTR) is clicks divided by impressions: the share of people who saw your result and clicked it. CTR rises and falls with position, so compare it at similar positions, or against a group of unchanged pages, when you test new titles and descriptions; a better CTR at a worse position isn’t a like-for-like win.
Content decay is the gradual loss of a page’s traffic and rankings as its facts age, competitors publish better pages, intent shifts or the results page changes, for example when AI answers absorb simple queries. It is slow, so it goes unnoticed: compare each page’s clicks with the same period last year in Search Console and review your content every quarter.
Core Web Vitals are Google’s three metrics for real-user page experience: LCP for loading, INP for responsiveness and CLS for visual stability. A page passes when, at the 75th percentile of visits, LCP is 2.5 seconds or less, INP is 200 ms or less and CLS is 0.1 or less; Google’s ranking systems use them, but relevance still comes first.
Crawl budget is the set of URLs Google can and wants to crawl on your site, shaped by how much load your server can take and how much demand Google sees for your pages. Google says it mainly matters for very large or fast-changing sites and that most sites needn’t worry about it, though filter and parameter URLs can turn even a mid-sized shop into a very large site in URL terms.
Crawling is how search engines discover and fetch URLs, using automated bots such as Googlebot that follow links and sitemaps. A page has to be crawled before it can be indexed, but crawling doesn’t guarantee indexing, and robots.txt controls what bots may fetch, not what appears in results.
Cumulative Layout Shift (CLS) measures how much visible content moves unexpectedly during the life of a page, scored as its largest burst of layout shifts. Good is 0.1 or less and poor is over 0.25 at the 75th percentile; the usual causes are images and ads without reserved space, banners injected above the content and late-loading web fonts.
Digital PR is earning coverage, mentions and links from online publications by giving journalists something worth reporting, such as original data, expert comment or a strong story. It earns links on merit and the kind of third-party coverage AI answers tend to draw on; publish a clear methodology and never stretch a finding further than the data goes.
Doorway pages are sites or pages created to rank for specific, similar searches, typically one page per town or keyword variant, each funnelling visitors to the same place. Google’s spam policies treat them as doorway abuse; a location page is legitimate when it serves a real branch or a genuinely covered area, with details that would be wrong if you swapped the city name.
Duplicate content is the same or very similar content at more than one URL, such as HTTP and HTTPS versions, filter and sort URLs or printer-friendly pages. Some duplication is normal and isn’t a breach of Google’s spam policies; the real cost is that signals split between versions, crawling is wasted and Google may show a version you didn’t choose, so consolidate with redirects or canonicals.
E-E-A-T stands for experience, expertise, authoritativeness and trust, a framework from Google’s Search Quality Rater Guidelines; “experience” was added in December 2022, and trust is the most important part. It isn’t a score or a ranking factor in itself: raters’ assessments don’t change rankings directly, but Google’s systems look for signals that line up with it.
An entity is a distinct, identifiable thing, such as a person, organisation, place or product, that search engines and AI systems recognise along with its attributes and relationships, as in Google’s Knowledge Graph. Describe your business the same way everywhere and add Organization structured data with sameAs links to your official profiles; contradictions make it harder for any system to be confident about who you are.
Faceted navigation is the set of filters, such as colour, size, brand or price, that shoppers use to narrow down a category, often creating a new URL for each combination. Google warns it can generate infinite URL spaces, so give landing pages only to facets with real search demand and enough products, and keep the rest out of the crawl with robots.txt rules or filter links that can’t be crawled.
A featured snippet is a result shown in a special box with the descriptive text first, answering the query above the usual title and link. Google chooses featured snippets automatically, and there is no markup to request one; the only guaranteed way to opt out is nosnippet, while max-snippet and data-nosnippet limit the text that can be shown.
Field data is performance data collected from real visitors on their own devices and connections, such as the Chrome UX Report data shown in PageSpeed Insights and Search Console’s Core Web Vitals report. It is what Google assesses, over a rolling 28-day window, so use it to decide what to fix and lab data to work out why; a fix only shows up in field data after a few weeks.
Generative engine optimisation (GEO) is work to make a brand visible and cited in AI-generated answers such as Google’s AI Overviews and AI Mode, ChatGPT search, Perplexity and Copilot. It builds on SEO rather than replacing it: let search crawlers in, write self-contained, answer-first passages, keep your entity details consistent and earn mentions elsewhere, since how these systems pick sources isn’t public.
Google Analytics 4 (GA4) is Google’s analytics platform, which took over when Universal Analytics stopped processing data in July 2024. It measures what visitors do on your site, but only those whose browser runs the tag and who consent where consent is required; standard properties keep event data for explorations for 2 months by default, or 14 if you change the setting, so change it early.
A Google Business Profile is a business’s free listing on Google Search and Maps, with its name, categories, address or service area, hours, photos and reviews. The primary category and how complete the profile is shape which local searches it shows up for; the name must be the real-world business name with no added keywords, and reviews must never be incentivised or requested only from happy customers.
Google Search Console is Google’s free tool for site owners: it shows how pages perform in Google Search (impressions, clicks, CTR and position), which URLs are indexed and why others aren’t, Core Web Vitals and rich result status. It keeps 16 months of performance data, so export a baseline before big changes; appearances in AI Overviews and AI Mode count within its normal Web data.
Googlebot is Google Search’s crawler, in two versions: Googlebot Smartphone, which makes most requests because Google indexes the mobile version of most sites, and Googlebot Desktop. Anyone can fake its user agent, so verify real Googlebot with a reverse DNS lookup or against the IP ranges Google publishes before you trust log data or write bot rules.
Helpful, people-first content is Google’s term for content made mainly to help people rather than to manipulate search rankings, set out in its guidance on creating helpful, reliable content. Google’s helpful content signals are now part of its core ranking systems, and warning signs include writing to a target word count, summarising others without adding anything and changing dates to look fresh.
hreflang is an annotation that tells Google which URLs are language or regional versions of the same page, using codes such as en-GB, en-US and tr-TR (never en-UK). It only works when it is reciprocal: each version must list every other version, ideally itself too, and point at indexable, self-canonical URLs. It doesn’t boost rankings; it helps Google show the right version in each market.
In Search Console, an impression is counted each time a link to your site appears in a result someone loads, usually whether or not it is scrolled into view. Appearances in AI Overviews and AI Mode count too, so impressions can rise while clicks stay flat; since June 2026 the separate Search Generative AI performance report shows those AI impressions on their own, and they are already part of your Web totals.
Indexing is when a search engine analyses a crawled page and stores it in its index, the database it draws results from. Not every crawled page is indexed: Google may judge it a duplicate or not useful enough, and Search Console’s Page indexing report shows this with statuses such as “Crawled – currently not indexed”.
IndexNow is an open protocol for telling search engines straight away that a URL has been added, updated or deleted, using a key file hosted on your site. A ping to one participating engine is shared with the others, which include Bing, Yandex, Seznam and Naver; Google doesn’t use it, and a ping is a hint, not a guarantee of crawling or indexing.
Interaction to Next Paint (INP) measures how quickly a page visibly responds to clicks, taps and key presses throughout a visit, and reports one of the slowest. It replaced First Input Delay (FID) as a Core Web Vital in March 2024; good is 200 ms or less and poor is over 500 ms at the 75th percentile, and long JavaScript tasks are the usual cause of failures.
Internal linking is how the pages of a site link to one another, through navigation, breadcrumbs and links within the content. It is how crawlers discover pages and one of the ways search engines judge which pages matter and what they’re about, so link to important pages prominently, with descriptive anchor text and real <a> elements that have an href.
JavaScript SEO is the work of making sure sites built with JavaScript frameworks can be crawled, rendered and indexed. Google can run JavaScript, but it renders after crawling and many other crawlers skip that step, so put main content, links, titles, canonicals and noindex rules in the initial HTML, use real links with an href and return real 404 status codes for missing pages.
JSON-LD is a format for adding structured data to a page as a block of code in a <script type="application/ld+json"> tag, kept separate from the visible HTML. Google recommends it because it is easy to generate and maintain; build it from the same data as the visible page, and on product pages put it in the initial HTML rather than injecting it with JavaScript.
A key event is an action you mark as important in GA4, such as an enquiry, call, booking or purchase; it is what GA4 used to call a conversion. In Google Ads, the key events you import become conversions. Mark only real business outcomes, not page views or scrolls, and report organic performance against them rather than against traffic alone.
Keyword cannibalisation is when two or more pages on your site compete for the same query and hold each other back, often visible as the ranking URL swapping between them. Two pages ranking together isn’t always a problem; when it is, merge them into the stronger URL with a 301, sharpen each page’s intent, or point internal anchor text at one page only.
Keyword research is finding the words and phrases your customers use when they search, with their volume, intent and competition, and deciding which page should target each. Group variations by intent instead of giving each its own page, use your own Search Console queries as well as tools, and research each market in its own language rather than translating keywords.
Lab data is performance data from a controlled test, such as Lighthouse or a DevTools recording, run on a device and connection you choose. It is repeatable and good for finding the cause of a problem, but it isn’t what Google assesses, and a standard page-load test can’t measure INP, which depends on real interactions; confirm every fix in field data.
Largest Contentful Paint (LCP) measures how long the largest image, text block or video in the viewport takes to appear, from the moment the visitor starts navigating to the page. Good is 2.5 seconds or less and poor is over 4 seconds, at the 75th percentile; the LCP element is often a hero or product image, so put it in the HTML, prioritise it and never lazy-load it.
Link reclamation is turning value you already have into links: asking sites that mention your brand or quote your data without linking to add a link, and getting links that point to broken, redirected or wrong pages corrected. It is often easier than earning new coverage, because the publisher has already chosen to feature you; a short, polite note pointing to the source page is often enough.
Link schemes are attempts to manipulate rankings with links, such as buying or selling links that pass ranking credit, excessive link exchanges, and keyword-rich links in widgets, footers or press releases spread across other sites. Google’s spam policies now call this link spam; it can lead to lower rankings or a manual action for unnatural links.
llms.txt is a proposed Markdown file at the root of a site that summarises its content for large language models. It is cheap to add, but no major AI search engine currently relies on it and Google has said it doesn’t use it, so treat it as optional and put the effort into crawler access and content that is easy to quote.
A local citation is any online mention of a business’s name, address and phone number, usually in directories, map apps, review sites and industry associations, such as Yell in the UK, Yelp in the US, Apple Business Connect or Bing Places. Focus on accuracy in the directories that matter in each market rather than on volume, and update every citation when you move or change number.
The local pack is the map with a short list of businesses, usually three, that Google shows for searches with local intent, such as “accountant near me”. Google ranks these results on relevance, distance and prominence and says you can’t pay for a better local ranking, so the work goes into your Business Profile, reviews, citations and links.
Localisation adapts a site for a market rather than just translating it: spelling and vocabulary, currency, prices with the right taxes, units, delivery terms, examples and proof. Start with keyword research in the local language, because translated keywords are a poor guide to how people search, and remember that Google judges language from visible content, so near-identical versions give it little reason to show the right one.
A location page is a page on your website for one branch or one genuinely served area, with its address, hours, map, directions, team, services and local reviews. Each Business Profile should link to its own location page; if you could swap the city name and nothing else on the page would be wrong, it is a doorway page rather than a location page.
Log-file analysis is the study of your server’s access logs, which record every request, to see how search engine and AI bots actually crawl the site: which URLs they request, how often and with what status codes. Verify Googlebot before you trust a line, compare the results with Search Console’s Crawl stats report, and treat raw logs as personal data, because they contain visitors’ IP addresses.
A meta description is a short summary of a page in a <meta name="description"> tag. Google doesn’t use it for ranking and often builds the snippet from page content that fits the query instead, but a clear, unique description still shapes clicks; there is no length limit, though snippets are cut to fit the screen.
NAP stands for name, address and phone number, the core details that identify a local business. They should match exactly on your website, your Business Profile and every directory that lists you; mismatches, often old addresses on forgotten directories, confuse customers and weaken the evidence that all those listings describe one business.
noindex is a rule, set in a robots <meta> tag or an X-Robots-Tag HTTP header, that tells search engines not to show a page in their results. It only works if the page can be crawled, so never block the same URL in robots.txt, and put it in the initial HTML: if Google finds noindex there it may skip rendering, so a script that removes it later won’t help.
An orphan page is a page that no other page on the same site links to. Search engines may still find it through a sitemap or an external link, but without internal links it is crawled less and receives little internal authority, so it rarely performs; compare a crawl of your site with your sitemaps, analytics and logs to find these pages, then link to them or retire them.
The Page indexing report in Google Search Console shows how many of your URLs are indexed and why the others aren’t, with statuses such as “Page with redirect”, “Soft 404” and “Crawled – currently not indexed”. That last one is usually a quality or duplication problem rather than a technical bug: Google fetched the page and decided not to index it.
A pillar page is the central page of a topic cluster: it gives an overview of a broad subject and links to supporting pages that cover each subtopic in depth, which link back to it. The common mistake is repeating the supporting pages’ content on the pillar, which leaves the pages competing for the same searches.
Query fan-out is the technique Google says AI Mode, and AI Overviews too, may use: breaking a question into subtopics, running searches for them at the same time and combining the results into one answer. It means your page can be retrieved for a related sub-question rather than the exact query, which rewards covering a topic thoroughly instead of writing one thin page per keyword.
A redirect chain is a series of redirects, such as A to B to C, that a URL passes through before reaching its final page. Google’s crawlers follow up to 10 hops, but Google says long chains hurt crawling and each hop adds a delay for visitors, so point internal links, sitemaps and older redirects straight at the final URL.
A referring domain is a website that links to yours at least once, counted once however many links it sends. It is a better measure of a link profile than the raw backlink count, because one site with thousands of footer or sitewide links is still a single source; track how many relevant referring domains you gain, not total links.
Rendering is running a page’s HTML, CSS and JavaScript to build what a visitor would see on screen. Google renders with an up-to-date version of Chromium, but only after crawling and sometimes with a delay, so content, links and tags that exist only after JavaScript runs can be picked up late; many other crawlers, including many AI crawlers, don’t run JavaScript at all.
Rich results are search results with extra features drawn from structured data, such as product prices and availability, review stars, breadcrumbs or event details. Valid markup makes a page eligible but guarantees nothing; Google stopped showing FAQ rich results altogether in May 2026 and retired HowTo results earlier, and businesses can’t earn stars by marking up reviews of themselves.
robots.txt is a plain-text file at the root of a site, such as example.com/robots.txt, that tells crawlers which URLs they may fetch. It controls crawling, not indexing: a blocked URL can still be indexed without its content if other pages link to it, and Google can’t see a noindex on a page it isn’t allowed to crawl, so use noindex or password protection to keep pages out of results.
Scaled content abuse is Google’s spam policy term for generating many pages mainly to manipulate search rankings rather than to help people, whether the text comes from AI, people, templates or a mix. Using AI isn’t the problem in itself; volume without value is, such as hundreds of near-identical location or keyword pages, and it can push a site down or out of results.
Schema.org is the shared vocabulary of types and properties, such as Organization, Product and LocalBusiness, that structured data is written in; it was founded by Google, Microsoft, Yahoo and Yandex. It is far bigger than what Google uses: only some types earn rich results, so check Google’s documentation, and use the Schema Markup Validator for general schema.org checks.
Search intent is what a searcher actually wants from a query: to learn something, reach a particular site, compare options or buy. Plan one page per intent rather than one per keyword, and check intent against the results page: if two queries return largely the same URLs, Google is treating them as the same need, and one page should serve both.
A SERP is a search engine results page: everything shown for a query, from ads and organic results to AI Overviews, local packs, featured snippets, videos and shopping results. Its layout changes with the query, location and device, so check what actually appears for your keywords before you plan a page or read a ranking report; position 1 can sit below several other features.
Server-side rendering (SSR) means the server builds a page’s complete HTML for each request, instead of sending a near-empty shell that JavaScript fills in inside the browser. Every crawler then gets the content and links straight away; static generation does the same at build time, while dynamic rendering, serving bots a separate pre-rendered version, is only a workaround in Google’s view.
A service-area business visits or delivers to customers rather than serving them at its own premises, such as a plumber, cleaner or mobile mechanic. On Google Business Profile it removes its address and sets up to 20 service areas, within about two hours’ drive of its base; a business that also receives customers is a hybrid and keeps its address visible.
A site migration is any change that moves or restructures URLs at scale: a redesign with new paths, a new platform, a change of domain or a merger of sites. Most of the risk comes from missing redirects and lost content, so Google recommends 301 redirects from every old URL to its new equivalent, kept for at least a year, plus the Change of Address tool for a new domain; expect rankings to fluctuate for a while.
A soft 404 is a page that says the content doesn’t exist, or is nearly empty, but returns a 200 success code instead of a 404 or 410. Google flags these in the Page indexing report and keeps crawling them, so return a real 404 or 410 or give the page genuine content; redirecting many old URLs to an irrelevant page such as the homepage can be treated as a soft 404 too.
rel=sponsored and rel=nofollow are link attributes that tell Google about your relationship with the page you link to: sponsored for ads and paid placements, ugc for user-generated links such as comments, and nofollow when neither applies. Paid links need sponsored (Google’s preferred value) or nofollow; without one of them, they breach Google’s spam policies.
Structured data is code that describes a page’s content in a standard vocabulary, usually schema.org written as JSON-LD, so search engines can tell that a number is a price or a date is when an event starts. It makes a page eligible for rich results but doesn’t guarantee them, isn’t a ranking factor in itself and must describe content visitors can see on the page.
A subfolder (or subdirectory) puts each market or language under a path on one domain, such as example.com/tr/ or example.com/us/. It shares the authority the domain has already earned and leaves one site to maintain, which makes it the usual default for international sites; pair it with reciprocal hreflang so Google knows which version is meant for which market.
Thin content is a page with little or no value of its own, such as a near-empty category, a filter combination with one product or a scraped or barely reworded article. Google can apply a manual action for “thin content with little or no added value”, and such pages often go unindexed; improve, merge or remove them rather than hiding them with noindex.
The title tag is the <title> element in a page’s HTML and the main source of the clickable headline in Google’s results, which Google calls the title link. Google rewrites titles that are too long, stuffed, boilerplate, duplicated or out of step with the page, and truncates them by pixel width (roughly 600 px on desktop) rather than by a fixed number of characters.
A topic cluster is a group of pages covering one subject: a pillar page for the broad topic and supporting pages that each answer one narrower question in depth, linked to each other. It shows readers and search engines which page covers what; built carelessly, with a page for every keyword variation, it turns into cannibalisation.
URL parameters are the name and value pairs after the ? in a URL, such as ?colour=black&sort=price, used for filtering, sorting, tracking and sessions. Each combination can count as a separate URL, multiplying duplicates and wasting crawling; Search Console’s URL Parameters tool has been removed, so control them on the site with links, robots.txt rules, canonicals and a fixed parameter order.
UTM parameters are tags added to a link, such as utm_source, utm_medium and utm_campaign, that tell GA4 where a visit came from. Use them on links you place elsewhere, such as emails, social posts and your Business Profile’s website link, whose visits otherwise blend into ordinary organic traffic; values are case-sensitive, so agree a lowercase naming scheme.
x-default is the hreflang value that names the fallback page for searchers whose language or region matches none of the versions you list, often a language selector or your international English page. It isn’t required, but if you use it, choose one URL and use it consistently across the whole set; pointing it at a page that redirects visitors by location defeats the purpose.
An XML sitemap is a file that lists the URLs you want search engines to find, optionally with the date each last changed. Include only canonical, indexable URLs that return 200; one file holds up to 50,000 URLs or 50 MB uncompressed, Google ignores priority and changefreq, and it uses lastmod only when the dates are consistently accurate.
YMYL, short for “your money or your life”, describes topics that could significantly affect someone’s health, finances, safety or wellbeing, such as medical, legal, tax or financial advice. Google says its systems give even more weight to strong E-E-A-T on these topics, so named, qualified authors, sources and clear company details matter most here.
A zero-click search is one that ends without a click to any website, because the results page answers it directly through an AI Overview, a featured snippet, a knowledge panel or a local pack. It isn’t always lost value: a call or a direction request from a Business Profile also happens without a click, so measure impressions, brand searches and conversions alongside clicks.
No terms match your search.
Get a free SEO audit
Send us your URL. Within 48 hours a senior strategist will send you a personalised audit with your biggest opportunities, completely free.