URL Structure

A URL is the address you type to land on a page, invented to replace IP addresses with something a human can remember. It is also the most compact description of a page that exists: read the path, and you should know what is on the other side of the click. That is the whole job of URL structure in SEO, and most of the work is decided once, when the site is built.

The parts break down like this:

What makes a URL SEO friendly

Three arguments carry the whole discipline.

Readability. A descriptive URL tells both people and machines what the page covers without them opening it. /restaurants/restaurant-copenhagen and /restaurants/249491.html?filter=cph lead to the same page. Only one of them earns the click when it appears in a search result, in a shared link or in a citation from an AI answer.

Relevance signals. As a ranking factor the URL is modest, and Google has said for years that keywords in a domain name are not one. But the path is used to understand what a page is about, so it is worth getting right. Note the order of importance: the path matters more than the domain. A site can rank perfectly well for "restaurant odense" on a domain with no keyword in it, as long as the category and page paths are clear.

Link structure. URLs express your hierarchy. Built in clear silos, they make the taxonomy and architecture of the site legible to crawlers, and they make internal linking obvious rather than improvised. Externally, a clean URL is easier for someone else to link to and quote.

A good default pattern: https://domain.tld/category-keyword/keyword. Follow it for products, articles and landing pages, and the structure stays predictable as the site grows.

Seven decisions worth making deliberately

1. HTTP or HTTPS

HTTPS, without discussion. It protects your users, it is a lightweight ranking signal, and browsers flag anything else as insecure. It requires an SSL certificate, which you get from your hosting or domain provider.

2. Subdomain or subfolder

Plenty of sites put the shop on shop.domain.com or the blog on blog.domain.com. It works, and Google treats subdomains fine. The practical argument for subfolders is simpler: domain.com/shop and domain.com/blog keep everything on one host, so authority, reporting and internal linking stay in one place instead of being split across properties you then have to reconcile.

3. Keywords in the URL

Descriptive words in the path help users and improve click-through rates. They are not a reason to buy a new domain. If your domain carries no keyword, put the effort into the paths instead, so each page states plainly what it covers.

4. Special characters and non-ASCII letters

Accented and Nordic characters look fine in an address bar and cause real trouble underneath. A domain like kolbergkød.dk becomes https://xn--kolbergkd-s8a.dk/ in practice, which is neither readable nor easy to share. Write the letters out instead, and keep &, #, ? and spaces out of paths entirely. Skip filler words like "the" and "and" while you are at it: they add length without adding meaning.

5. Country domain or .com

A country domain can help locally, since users in a market often trust their own extension. A .com with localized subfolders is easier to expand into new markets later. Neither choice is wrong, but changing your mind afterwards means a full migration, so decide it against your growth plan rather than by preference. If several markets are already on the roadmap, our guide to international SEO covers the folder and hreflang side of that decision.

6. Word separation

Use hyphens. Search engines read hyphens as word breaks and underscores as part of a single word, so red-widgets is three signals and red_widgets is one. Keep paths short while they stay descriptive, and keep them lowercase, since some servers treat uppercase and lowercase as different URLs.

7. Trailing slashes

Technically, /page/ and /page can be two different URLs, and a server may treat the first as a directory and the second as a file. Search engines can end up with both versions of the same content.

The fix is consistency rather than a correct answer. Pick one convention for the whole site, redirect the other version with a 301, and make sure your canonical tags point to the version you chose. Then use the same form in every internal link and in your sitemap, so you are not sending crawlers through a redirect on every click.

Changing URLs on a live site

Old URLs that are unreadable can be fixed, and the process is well understood. Two steps carry it:

1. Redirect with a 301. A permanent redirect tells search engines and users that the page has moved, and passes the accumulated value from the old address to the new one. Redirect chains and loops are where this goes wrong, so map old to new directly and verify each one. The mechanics live in site health and redirects.
2. Update your internal links. Change every internal link to point at the new URL rather than relying on the redirect. This is the step teams skip, and it is the one that keeps crawl paths clean.

Then make sure the new URLs are in your XML sitemap. A URL that is not listed there is a URL you have not asked to have found, and everything about how it gets discovered runs through crawling and indexing.

Changing URLs always carries some risk, so it is worth doing once and doing it properly. If you want a read on how your current structure holds up, along with the rest of your technical foundation, a free SEO analysis covers it.