Somebody pastes a link into a message. Before anyone clicks it, the URL has already said something about where it leads.
`/services/sustainable-web-design/` is legible. `/index.php?p=447&cat=12` is not. The page behind either could be identical, but only one tells you what you are about to open, and that difference matters to people, to search engines and increasingly to systems reading your site without a browser.
URL structure in SEO is rarely the thing holding a site back. It is, however, cheap to get right at the start and genuinely awkward to fix later.
URL Structure in SEO: Semantic Signposting for 2026
What a good URL does
- Describes the content in words. Real words, readable aloud, matching how people would describe the page.
- Shows where it sits. `/services/sustainable-web-design/` conveys a hierarchy. A flat structure where everything hangs off the root loses that information.
- Stays short. Long URLs get truncated in results, break awkwardly when shared and are harder to read.
- Uses hyphens between words. Underscores are read differently and spaces become `%20`, which is unreadable.
- Stays lowercase. Some servers treat cases as different addresses, which quietly creates duplicate URLs for the same page.
- Omits dates unless the content is genuinely time-bound. A date in the URL of an evergreen article makes it look stale long before it is.
What it does not do
It does not rank pages. Keywords in URLs are a minor signal at most, and stuffing them there is a recognisable pattern rather than an advantage.
Anyone proposing a URL restructure primarily for ranking benefit is proposing significant risk for marginal return. The reasons to get URLs right are clarity, maintainability and not confusing your own team, all of which matter more than the search signal.
Changing existing URLs
This is the part that goes wrong, so it deserves the most space.
Every URL that has existed for a while has accumulated something: links from other sites, bookmarks, references in emails, entries in search indexes. Changing it without redirecting discards all of that.
- Redirect every old URL to its closest equivalent with a 301, which tells search engines the move is permanent and passes the accumulated signals across.
- Never redirect everything to the homepage. It is treated as a soft failure, and it is a poor experience for someone expecting a specific page.
- Update your internal links to point at the new URLs directly. Redirect chains slow things down and dilute the signal at each hop.
- Keep redirects in place indefinitely. Removing them a year later breaks links that still exist elsewhere on the internet.
- Expect a temporary dip while search engines reprocess. That is normal, and panicking mid-migration by changing things again is how a temporary dip becomes a permanent one.
Google’s documentation on redirects covers the mechanics.
Canonical tags for the duplicates you cannot avoid
Some duplication is unavoidable: filtered listings, tracking parameters, printable versions. A canonical tag tells search engines which version to treat as authoritative.
The common failure is a canonical tag pointing somewhere unintended, usually left over from a template or a staging environment. That instructs search engines to ignore the page entirely, and it happens more often than people expect.
Check your important pages actually canonicalise to themselves.
Trailing slashes and other small consistency problems
`/services/` and `/services` can be treated as different addresses. So can `http` and `https`, and `www` and non-`www`.
Pick one form for each, redirect the others to it, and be consistent in your internal links. None of this is difficult. All of it quietly creates duplicate URLs when ignored.
Where performance and environmental impact both apply, treat them as the same lever: reducing unnecessary data transfer and processing supports a faster experience while lowering estimated digital carbon, though environmental figures remain modelled rather than directly measured. Redirect chains are a small but real instance of this, adding round trips to every affected request.
Our search engine optimisation work covers URL structure as part of technical health rather than as a project of its own.