Websites & Hosting

What Is a Website? A Plain Answer

Not a definition copied from a dictionary — what a website is made of, what has to exist for one to work, and why "website" covers things as different as a blog and a bank.

FundamentalsHostingBeginners

"Website" is one of those words that gets used constantly and explained rarely. Most definitions are circular — a website is a site on the web — which is true and useless.

Here's a more useful version, and the rest of this guide unpacks it: a website is a collection of documents, stored on a computer that stays switched on, which any browser can request over the internet using one memorable name. Every part of that sentence is doing work.

The short definition

Break the sentence into its claims:

  • A collection of documents. Usually pages of text and images, linked to each other. A single page is technically a website; most have many.
  • Stored on a computer that stays switched on. That computer is a server. "Stays switched on" is the entire point — if it sleeps, the site is unreachable.
  • Any browser can request them. The documents aren't sent anywhere. They sit and wait, and are handed over on request, one visitor at a time.
  • Using one memorable name. The domain. Without it you'd need to remember a numeric address instead.

Miss any one of those and you don't have a working website. Files on your laptop aren't a website — nothing can reach them. A domain with nothing behind it isn't a website either.

The three things a website needs

These are separate purchases, often from separate companies, and confusing them is the most common beginner mistake:

  • A domain — the name people type. You rent it annually from a registrar. It does not include storage or a website.
  • Hosting — space on a server that stays online and answers requests. Rented separately, often monthly or annually.
  • The site itself — the actual pages. Built by you, by someone you pay, or generated by software you install.
Note

Buying a domain does not create a website, and it does not include hosting. Registrars sell hosting too, which is why the two get conflated — but they're distinct things, and you can buy them from different companies.

How these three connect — how a typed name reaches the right server — is the subject of How Websites Actually Work.

What a page is actually made of

Open any page, and three kinds of file are usually involved:

  • HTML — the content and its structure. Headings, paragraphs, lists, links, form fields. This is the page; the other two are optional.
  • CSS — how it looks. Typefaces, colour, spacing, layout, and how the layout changes on a narrow screen.
  • JavaScript — behaviour that responds to the visitor. Optional, and worth being deliberate about; see What Actually Makes a Web Page Interactive.

A page with only HTML works. It will look plain, but it will load, be readable, and be navigable. That's worth knowing: the content layer is the one that genuinely can't be removed.

Fixed pages and generated pages

Two ways a server can answer a request, and the difference explains a lot about how sites are built.

Fixed (static) — the HTML file already exists. The server finds it and sends it, unchanged, to everyone. Fast, cheap, hard to break, and completely adequate for a brochure site, documentation, or a blog.

Generated (dynamic) — the HTML doesn't exist until it's asked for. Code runs on the server, usually fetching from a database, and assembles the page per request. Necessary when the page differs by visitor or changes often: a logged-in account, a cart, search results, anything personalised.

Neither is better. The question is whether the page needs to differ per visitor or per moment. If it doesn't, generating it fresh every time is work nobody benefits from.

The same word covers very different things

A handful of genuinely distinct types, rather than an exhaustive list:

  • Brochure sites — information about a business. Mostly fixed pages, rarely changed.
  • Blogs and publications — dated entries, newest first, organised by subject.
  • Documentation and reference — structured, searchable, updated alongside whatever it documents.
  • Shops — catalogue, cart, payment. Necessarily dynamic; handling money raises the security stakes considerably.
  • Applications — sites that are really software: mail, banking, project tools. The line between "website" and "app" is genuinely blurry here.

They differ enormously in complexity but share the mechanism: a browser asks, a server answers with HTML.

What a website isn't

Three things regularly mistaken for one:

  • A social media profile. You publish there; you don't control it. The rules, the reach, and its continued existence are someone else's decision.
  • A domain on its own. A name pointing at nothing.
  • A design file. A picture of a website isn't a website until it's built and served.

The distinction that matters most is the first: the reason to have a website, rather than only a profile somewhere, is that nobody can change the rules on you or take it away.

The notes

Practical web notes, without the noise.

Occasional useful notes about web development, websites, tools and workflows.

No daily emails. No hype. Unsubscribe whenever you like.

Sign-up isn't open yet.
These notes aren't being sent to a mailing list at the moment. Nothing to enter here yet — new guides appear on Learn as they're published.