Technical SEO and web performance glossary

Technical SEO, Core Web Vitals, performance and WCAG accessibility terms. It does not cover marketing SEO (links, keywords, authority): this is the technical layer, the one that can actually be measured. Where a real threshold exists, it is in the definition.

Looking for UX research and psychology terms? UX glossary

A

AI crawlability

How accessible a site's content is to bots from AI systems such as ChatGPT, Claude, Gemini or Perplexity. Unlike traditional search engines, which index in order to show results, these crawlers gather information to train models or to answer in real time. Deciding which ones to allow, via robots.txt, is an editorial decision as much as a technical one.

Alt text

A textual description of an image, written in the alt attribute, for those who cannot see it: screen reader users, people with images disabled and crawlers. WCAG 2.2 success criterion 1.1.1, Non-text Content, requires it at Level A. The text depends on the image's role: informative images are described, functional ones state the action, decorative ones take an empty alt.

ARIA labels

Attributes such as aria-label and aria-labelledby that give an accessible name to elements where native HTML provides none, like icon-only buttons or custom widgets. WCAG success criterion 4.1.2 requires every UI component to have a programmatically determinable name. Without one, the screen reader announces only button or link. Native HTML is preferred over ARIA.

C

Canonical URL (Canonical URL (rel="canonical"))

A link tag in the head that declares which version of a page is the preferred one when several URLs show identical or near-identical content, the typical case of filter and sort parameters in ecommerce. It is a strong signal, not a directive: Google consolidates signals onto that URL and usually shows only it, but reserves the right to pick a different one. Even so, the variants stop competing with each other.

Colour contrast (Colour contrast ratio)

The luminance difference between text and its background, expressed on a scale from 1:1 to 21:1. WCAG 2.2 success criterion 1.4.3 requires at Level AA a minimum ratio of 4.5:1 for normal text and 3:1 for large text, defined as 18 point or 14 point bold. A separate criterion, 1.4.11, requires 3:1 for interface components and graphical objects. Level AAA (criterion 1.4.6) raises normal text to 7:1.

Critical CSS

The minimum subset of styles needed to render above-the-fold content, inlined in the HTML while the rest of the stylesheet is deferred. The browser cannot build the render tree until it has processed all CSS found in the head, so a 200 KB stylesheet delays painting even when only 8 KB cover what the user actually sees.

Cumulative Layout Shift (CLS)

Quantifies visual stability by summing the unexpected layout shifts that happen over the life of the page. Each shift is calculated by multiplying the fraction of the viewport affected by the distance moved. Good is up to 0.1, needs improvement between 0.1 and 0.25, and poor above 0.25. Shifts triggered by user interaction do not count.

E

Explicit image dimensions (Explicit image dimensions (width/height))

Declaring width and height attributes on every img tag so the browser reserves the space before the image finishes downloading. Without them the browser collapses that area to zero and pushes content around when the image loads. It is the most frequent cause of Cumulative Layout Shift, and the practice was lost to responsive design before returning for that reason.

F

Focus indicators

The visual cue showing which interactive element currently has keyboard focus. WCAG success criterion 2.4.7 requires it to be visible. Browsers ship a default, styled differently across Chrome, Firefox and Safari. If you remove it with outline: none and do not replace it, keyboard users lose track of where they are on the page.

H

Heading hierarchy (Heading hierarchy (H1–H6))

The structure that organises content with H1 to H6 tags, where each level signals a section's importance relative to the one above. H1 is the page's visible title; H2 opens main sections and H3 subdivides them. Screen readers use this hierarchy to jump between sections, so skipping levels breaks that navigation.

hreflang

An attribute that declares which language or regional version of a page matches each user, so equivalent translations are not treated as duplicate content. You write one link tag per version, including a self-reference, using ISO 639-1 language codes and optional ISO 3166-1 region codes. It requires bidirectional return links; without them Google ignores the annotations.

HTTP caching headers

Response headers, led by Cache-Control, that tell the browser whether it may store a resource locally and for how long. Configured well, they avoid re-downloading files on later visits and can cut load times by 50 to 90% for returning visitors, while also reducing server load and bandwidth costs.

I

K

Keyboard navigation

The ability to use an entire site with the keyboard alone, no mouse or touchscreen. WCAG success criterion 2.1.1 requires all functionality to be available from a keyboard. People with motor disabilities, screen reader users and alternative devices such as switches, sip-and-puff and eye tracking all depend on it, since those devices emulate keyboard input.

L

lang attribute

An HTML attribute declaring the human language of the content. WCAG success criterion 3.1.1, Level A, requires each page's default language to be programmatically determinable: the html element must carry lang with a valid BCP 47 language tag. Screen readers use it to pick the right voice and pronunciation rules; without it they may read text with the wrong phonetics.

llms.txt

A plain text file proposed as a standard, placed at the site root, that gives AI systems a structured view of the content: a concise description of the site, links to its most important pages and how the content is organised. It works as a README for machines. It is a proposal, not a ratified standard, and adoption across providers is still uneven.

M

Meta description

A meta attribute in the head that summarises a page's content and usually appears under the title in search results. It is not a direct ranking factor; it influences CTR. Google publishes no length limit and truncates the snippet to fit the device width; in practice you see about 155 to 160 characters on desktop and 120 to 130 on mobile, so keeping it between 120 and 155 avoids truncation on either.

Meta robots tag

A head tag that tells search engines how to treat a specific page. Its most used directives are noindex, which keeps it out of results, nofollow, which stops link following, and nosnippet, max-snippet or max-image-preview to control the preview. The X-Robots-Tag HTTP header does the same for non-HTML resources such as PDFs or images.

Modern image formats (Modern image formats (WebP, AVIF))

WebP and AVIF use compression algorithms derived from video codecs to produce smaller files at equivalent visual quality. WebP, from Google, relies on VP8 predictive coding for lossy compression. By Google's own figures, lossy WebP is 25 to 35% smaller than an equivalent-quality JPEG, so serving only JPEG and PNG usually means sending considerably more data than necessary, with the corresponding cost in load time.

R

Render-blocking resources

Files, mainly CSS and JavaScript in the head, that stop the browser from showing anything until they are fully downloaded and processed. They interrupt the critical rendering path between HTML parsing and first paint, so they delay FCP, LCP and Speed Index even when the rest of the page is fast. You fix them by deferring, loading asynchronously, or inlining what is essential.

Responsive images (Responsive images (srcset))

A set of HTML attributes, mainly srcset and sizes, that offer several versions of the same image so the browser can pick the right one for the device, viewport width and screen density. Serving a 2000 px image to a phone with a 400 px viewport wastes bandwidth. Oversized images are one of the most common causes of poor mobile Core Web Vitals.

robots.txt

A plain text file at the domain root that follows the Robots Exclusion Protocol and tells crawlers which paths they may visit. It uses User-agent, Disallow, Allow and Sitemap directives. It controls crawling, not indexing: a URL blocked here can still appear in results if other pages link to it. To prevent indexing you need meta robots.

S

Semantic HTML

Using each HTML element for its intended purpose rather than its appearance: header, nav, main, article, lists and headings instead of div and span for everything. WCAG success criterion 1.3.1, Info and Relationships, requires structure to be programmatically determinable. When everything is a div, assistive technology loses landmarks and heading navigation.

Structured data (Structured data (schema.org))

A standardised schema.org vocabulary added to the HTML, usually as JSON-LD, to explicitly describe what a page's content is: an article, a product, a recipe, who created it and what properties it has. Without it, the search engine infers meaning from indirect signals. It enables rich results and underpins how AI systems extract verifiable facts.

T

Text compression (Text compression (Gzip, Brotli))

Compressing text resources such as HTML, CSS, JavaScript, JSON and XML before sending them. The browser declares what it supports in Accept-Encoding and the server responds naming the method in Content-Encoding. It cuts transfer size by 60 to 90%, which improves TTFB and FCP. Gzip has universal support; Brotli is newer and compresses harder.

Text resize (Text resize (200% zoom))

WCAG success criterion 1.4.4, Level AA, requires text to scale up to 200% without assistive technology and without loss of content or functionality. Nothing may be clipped, overlapped or hidden, and horizontal scrolling should not be needed at standard viewports. Captions and images of text are excluded. Fixed pixel sizing is the usual cause of failure.

Title tag

An HTML element inside the head that defines a page's title. It appears as the clickable link in search results, as the browser tab name and as the bookmark name. Google publishes no limit: in practice it shows roughly 50 to 60 characters, about 600 pixels on desktop, and truncates the rest by screen width, not an exact character count. It also frequently rewrites the title when it judges another one describes the page better.

Total Blocking Time (TBT)

Sums the time the main thread was blocked between FCP and Time to Interactive. Every long task, meaning any task over 50 milliseconds, contributes the excess above those 50 ms. In mobile Lighthouse, good is under 200 milliseconds, needs improvement between 200 and 600, and poor above 600 (desktop cutoffs are lower). While the thread is blocked, clicks do not register.

Touch target size

The clickable or tappable zone of a control, comprising the visible element plus any padding that extends the hit area. WCAG 2.2 success criterion 2.5.8, new in this version, requires at Level AA at least 24 by 24 CSS pixels, with exceptions for spacing, equivalents and inline text. Criterion 2.5.5 raises the minimum to 44 by 44 pixels at Level AAA.

V

Viewport meta tag

A head tag that tells the mobile browser how to fit the page's dimensions and scaling to the actual screen size. Its usual form is width=device-width, initial-scale=1.0. Without it, mobile browsers render the page at desktop width and scale it down, leaving text tiny and navigation awkward. It is the foundation of responsive design.

W

X

XML sitemap

An XML file listing a site's important URLs so search engines can discover them. It is aimed at machines, not people, and differs from the navigable HTML sitemap. It acts as an entry point for crawling and helps most with new content or pages with few internal links. It does not guarantee indexing, it only eases discovery.

Need to apply this to a real product?

A glossary explains the terms; deciding which one applies to your case is a different conversation. If you want to think it through with someone, get in touch.