Glossary

WWW vs Non-WWW SEO: Which Should You Use?

The www and non-www versions of a site are two different URLs — and hosts — until you pick one as canonical and redirect the other.

· By Rogier Bruggeman, Founder of KinetixSEO

RB
Rogier BruggemanFounder of KinetixSEO · 3 min read

What www vs non-www means for SEO

The www and non-www (bare domain) versions of a site's address are technically two different URLs — and, without configuration, two different hosts — so search engines and browsers treat www.example.com and example.com as separate entities unless the site owner picks one as canonical and redirects the other. Nothing about DNS or HTTP guarantees these two addresses point to the same place. A site can serve completely different content, or no content at all, from one of them. SEO only enters the picture because if both versions resolve and return the same page without a declared preference, Google can index both, split backlink and internal-link signal between them, and occasionally show the "wrong" one in search results.

Check your own site

Want to see this on your own site?

Use your own site as the evidence. Get a free SEO and AI-citation readiness baseline, then monitor what changes.

Why linking to both versions splits ranking signal

A site that inconsistently links to both versions internally — some navigation pointing to https://www.example.com/page, some to https://example.com/page — splits its ranking signal between two versions of every single page it does this on. Each inbound and internal link is a vote, and until the site consolidates, that vote gets divided rather than counted once. The fix is a 301 redirect from the non-canonical host to the canonical one, paired with a self-referencing canonical tag on the version you keep, so both crawlers and link equity converge on a single URL. Neither step alone is reliable: a redirect without a matching canonical still lets a crawler index the redirect target under conflicting signals if something elsewhere declares otherwise, and a canonical tag without a redirect leaves both hosts live and crawlable.

This isn't a theoretical risk. A KinetixSEO crawl of 36 public SEO and web-performance guide articles from software vendors and developer publications, run October 6, 2026, found that 3% of them point rel=canonical at a URL other than the one actually served after redirects — meaning even pages written to explain correct canonicalization aren't immune to the mismatch they describe (KinetixSEO crawl study). By contrast, a canonical-conflict check KinetixSEO ran on October 10, 2026 against its own hreflang tag page came back clean: self-referencing or no canonical, no mismatch between the declared and served URL.

www or non-www: which one should you pick for SEO

Neither version has an inherent ranking advantage — Google has said for years that www and non-www are equally valid, so the choice is about consistency, not SEO performance. Pick one host as canonical based on practical factors: existing backlink patterns (which version already has more external links pointing at it), cookie-scoping needs (www subdomains can scope cookies differently than bare domains, which matters for some app setups), and whatever your CDN or hosting provider defaults to with the least friction. Once you pick, apply it everywhere: the 301 redirect, the canonical tag, your internal links, your XML sitemap, and any hardcoded URLs in your CMS. For the mechanics of each piece, see the glossary entries on the canonical tag and 301 redirect — the index of consolidation techniques covers both in more depth, including how they interact with crawlability and indexability more broadly.

Consolidation also matters beyond traditional search. When KinetixSEO tracked an AI citation prompt on October 10, 2026 asking Gemini which tools track brand visibility across AI assistants, the answer cited 18 sources — including unite.ai, semseo-solutions.com, businessinsider.de, useomnia.com and viali.ai — and kinetixseo.com was not among them (KinetixSEO GEO citation tracking). Split signal from unresolved host duplication is exactly the kind of structural issue that can keep a domain out of results like that one, in search and in AI answer engines alike.

Sources

  1. KinetixSEO crawl study ()

    3% of the 36 public SEO and web-performance guide articles from software vendors and developer publications point rel=canonical at a URL other than the one they serve. Sample: 36 observations, measured Oct 6, 2026. Methodology: Fetched each of the 36 public SEO and web-performance guide articles from software vendors and developer publications, following redirects, and counted the pages whose rel=canonical — in the HTML head or the HTTP Link header — names a URL other than the one finally served. A page declaring no canonical at all is counted as self-referencing, not as a conflict.

  2. KinetixSEO GEO citation tracking ()

    On 10 October 2026, Gemini (web search) answered the tracked prompt "I'm an SEO consultant. Clients keep asking how visible they are in AI answers. What tools track brand mentions and citations across AI assistants like ChatGPT and Perplexity? Name specific products." and cited 18 sources: unite.ai, semseo-solutions.com, businessinsider.de, useomnia.com, viali.ai and 13 more. kinetixseo.com was not among them.

  3. KinetixSEO crawl check ()

    On 10 October 2026, KinetixSEO ran its "canonical conflict" check against https://kinetixseo.com/learn/hreflang-tag. What it tests: Whether the page's rel=canonical names a URL other than the one finally served, after redirects. Result: Self-referencing or no canonical.

Check your own site

See how your own site scores on SEO and AI-search visibility — free report, no signup.

Use your own site as the evidence. Get a free SEO and AI-citation readiness baseline, then monitor what changes.