A URL, short for Uniform Resource Locator, is the complete address of something on the web: a page, an image, a PDF. It is what appears in the browser's address bar, such as https://www.example.com/services/roof-repair?ref=email#prices. Every page on your site has its own URL, and that is how people, links and search engines find it.
The parts of a URL, left to right
- Scheme (
https://): the protocol. HTTPS means the connection is encrypted. - Host (
www.example.com): the subdomainwwwplus the domain name. DNS turns this into a server address. - Port: normally left out, since the scheme already implies one (443 for HTTPS, 80 for HTTP).
- Path (
/services/roof-repair): which page on that server. The last segment is often called the slug. - Query string (
?ref=email): extra key-value data, used for search filters, tracking and UTM parameters. - Fragment (
#prices): jumps to a section of the page. It is handled by the browser and never sent to the server.
URLs can only contain a limited set of characters, so spaces and many symbols are percent-encoded: a space becomes %20.
A real example of URL choices
A roofing company's old site used addresses like /page.php?id=47. On the new site the same page lives at /services/roof-repair. The new version tells a visitor what the page is before they click, reads well when shared in a text message, and gives Google a small hint about the topic. Because the address changed, the old one gets a 301 redirect to the new one so existing links and rankings carry over.
When URLs matter for your site
For most small sites you set URL structure once, at launch, and then leave it alone. The decisions that matter: short readable paths, lowercase letters, hyphens between words, and one consistent version of the domain (with or without www, always HTTPS) that the others redirect to. After launch, the biggest risk is changing URLs carelessly.
URL mistakes that cause real problems
- Changing URLs without redirects, which breaks bookmarks and backlinks and loses search visibility.
- Several addresses for one page, like with and without a trailing slash, splitting signals unless a canonical tag picks one.
- Dates or IDs in paths that make content look old or force URL changes later.
- Putting personal data in query strings, where it ends up in logs and analytics.
Related terms
- Slug: the readable final part of the path.
- Permalink: a URL meant never to change.
- Domain name: the host portion of every URL.
- UTM parameters: query-string tags for campaign tracking.