In short
A strong URL is short, readable, uses hyphens not underscores, reflects a logical folder hierarchy, and never needs to change — pick the structure once, get it right, and treat any later change as a real migration with redirects.
What makes a URL SEO-friendly
A good URL is descriptive enough that a person can guess what's on the page before clicking, and stable enough that it never needs to change. /insights/technical-seo-for-modern-websites/ tells you exactly what you're getting. /p?id=4821&cat=12 tells you nothing and depends on database internals that shouldn't be public.
The core rules
- Use hyphens, not underscores, to separate words. Google treats hyphens as word separators and underscores as joiners —
technical-seo-guidereads as three words;technical_seo_guidedoesn't reliably. - Keep it lowercase. Mixed-case URLs create duplicate-content risk on servers that treat paths case-sensitively.
- Keep it short and readable. Trim filler words ("the," "a," "of") unless they're needed for clarity.
- Reflect real hierarchy.
/insights/for articles,/capabilities/for services — the folder structure should match how a visitor (and a crawler) would categorize the content, not an arbitrary CMS default. - Avoid parameters for canonical content. Query strings are fine for filtering or tracking, but the canonical version of a page should have a clean path.
Matching structure to site architecture
URL structure isn't just cosmetic — it should mirror your actual information architecture. If /insights/ and /capabilities/ are your two main content types, every article and every service page should live under the matching folder, consistently, so both users and crawlers can predict where related content lives.
Changing URLs without losing rankings
If a URL genuinely needs to change — a rename, a restructure — treat it as a migration: set up a 301 redirect from the old path to the new one, update all internal links to point directly at the new URL (don't rely on the redirect chain), and update the sitemap. A single, correctly-configured 301 preserves the vast majority of a page's accumulated ranking signal; leaving the redirect out, or chaining multiple redirects, loses more of it.
What to avoid
- Session IDs or tracking parameters in canonical URLs — they create infinite duplicate-content variations.
- Overly deep nesting — a URL five folders deep is harder to remember, share, and often signals the content is buried too far from the homepage.
- Keyword stuffing —
/best-cheap-affordable-seo-services-2026/reads as spam to both people and ranking systems; one clear phrase beats several jammed together. - Changing structure repeatedly — every change costs some ranking signal even with a correct redirect; get the structure right early rather than iterating on it constantly.
A quick self-audit
Pick five pages at random and read only their URLs, with no other context. If you can't tell what each page is about from the path alone, the structure needs work.
Keep reading: How to Structure an Article for Featured Snippets · How to Add FAQ Schema That Wins Rich Results.
Common questions
Do URL keywords still matter for rankings?
They're a minor, not major, signal — but a clear, descriptive URL still helps both search engines and users understand the page, and it shows in the search result display.
Should I use dates in blog post URLs?
Generally no, for evergreen content — a dated URL (/2023/03/post-title/) can make a still-relevant article look stale, and it complicates updates. Reserve dates for genuinely time-bound content.
What happens if I don't redirect an old URL after changing it?
You lose most of the accumulated ranking signal for that page and anyone with the old link (bookmarks, backlinks, social shares) hits a 404 instead of your content.