Glossary

A copy of a site and its content kept somewhere other than the server it came from, recent enough that losing the gap would not hurt, and restored at least once to prove it works. A backup nobody has restored is a hope. Our sites keep every version of every page in version control in the client’s own account.

The step that turns a site’s content and templates into the finished pages that are served to visitors. On a static site it runs whenever the content changes, takes a minute or two, and produces every page in advance. It is also where we measure the site and generate its sitemap and headers.

Content delivery network: copies of a site’s files held on servers around the world, so a visitor in Texas and a visitor in Melbourne each get the page from somewhere near them. A static site is served entirely from one. It makes a good page fast everywhere; it cannot make a heavy page light.

Content management system: the editor a team uses to change what a website says without touching code. WordPress is one. The sites we build use a simpler editor, shaped around the content, that saves each change with a history.

Three measurements Google takes from real visits to a page: how long until the main content appears, how quickly the page responds to a tap, and how much it jumps about while loading. Google calls a page good when it passes all three for the slowest quarter of its visitors. They are the measure behind the word fast on this site.

What a search engine, or an audit tool, does when it follows every link on a site and reads every page. A crawl finds broken links, duplicate addresses and pages that cannot be indexed. The Website Risk Analysis starts with one.

A collection of technical documents, usually PDFs, each tied to a product, that buyers download to specify or compare. On a supplier’s site the library is often the most used part. Engineers rate data sheets the most valuable content on a supplier’s site, so the library’s structure, search and addresses matter more than the homepage.

A website whose pages are built by a program on the server, from a database, each time someone asks for one. WordPress works this way. It suits pages that differ for every visitor or change by the minute, such as accounts, live stock and search across a very large catalogue. The price is more running parts to attack and keep patched, and a page that is slower unless extra work is done to speed it up. See Static site for the alternative.

A way of building a website where the editor and the content are separate from the pages that visitors see. The content can then be served to a website, an app or a catalogue without being retyped. Our sites work this way, with the editor writing plain files and the build turning them into pages.

Google’s record of the pages it knows about and may show in results. A page that is not indexed cannot be found by search, however good it is. Whether a page is indexable depends on its address, its robots rules and whether its content is in the page rather than fetched by script.

Google’s testing tool, which loads a page once on a simulated slow mobile phone and scores it from 0 to 100 for performance, accessibility, general code quality and search readiness. It runs inside PageSpeed Insights and the Chrome browser. Useful for finding what to fix; not the same as passing Core Web Vitals, which come from real visits.

A piece of software added to a website system such as WordPress to give it a feature it did not have: a form, a gallery, a shop. Each plugin is written by someone else, runs on your server, and needs updating whenever its author fixes a flaw. In 2025 nine in ten of the security flaws disclosed in WordPress were in plugins.

An instruction that sends a visitor, and a search engine, from an old web address to a new one. A permanent redirect tells Google the new page is the one to index, so the old address keeps its standing. When we move a site, every old address is kept or redirected, and the list is shown before launch.

A file a website publishes listing every page it wants search engines to know about, with the date each last changed. Search engines read it to find pages quickly. Our sites generate it at every build so it cannot go out of date.

A website whose pages are made in advance and served as plain files, rather than built by a program on the server each time someone visits. There is nothing running on the public server to attack and nothing to wait for, which is why static sites are fast and have little to patch. The content is still edited, in an editor, and the site is rebuilt when it changes. It cannot show a different page to each visitor on its own; anything that must, such as signing in or paying, runs as a small separate service. See Dynamic site for the alternative.

Facts in a page written in a form machines read without guessing: this is a product, this is its code, this is its data sheet. Search engines use it to show richer results. It helps when it repeats what the page already says in plain text, and it does nothing on its own.

The encryption between a visitor’s browser and a website, shown by the padlock and the https at the start of the address. Without it anyone on the path can read or alter the page. Every site we build uses current versions only and tells the browser never to connect without it.