No Www

No www is the bare domain — example.com without the www. prefix. Search engines treat it as a separate site from www.example.com unless you tell them otherwise. I once consulted for a site that had been running for three years with both versions resolving to the same content. They had 150 backlinks split roughly evenly between the two, meaning each page carried half the authority it should have. After consolidating via [301 redirect](/301-redirect/) and updating their [canonical tags](/canonical-tag/), their organic traffic climbed 12% over the next three months. The lesson: choosing is trivial, enforcing is everything. You also need to update internal links — many CMS platforms have a 'site URL' setting that I forget to check. And if you use [duplicate content](/duplicate-content/) detection tools, they will flag the version mismatch until you fix it. For most small sites, no www is fine because it is shorter and easier to type. But www gives you a subdomain that can isolate cookies, which matters if you serve content across multiple subdomains. My decision rule: if you run a single blog or brochure site, go no www. If you have user logins, a CDN, or subdomains for different services, pick www and set the cookie domain carefully. Either way, the redirect is non-negotiable.

It matters most at launch. A new site without redirects is clean. But I have inherited sites where someone set up the staging environment on www.staging.example.com and then pushed that live, leaving the main domain on non-www. That takes an hour of [technical SEO](/technical-seo/) audit work every time you run a crawl. Another scenario: after a site migration or redesign, your redirect map must include the version switch. I had a client who changed from www to non-www but forgot to redirect the old www pages to the new non-www equivalents — they lost 40% of their indexed pages for a month. If you see duplicate content warnings in Search Console, check both versions. One of my sites had 18% of crawl budget wasted on the non-preferred version before I fixed it. And when building backlinks, always send people to the version you control. A competitor once pointed links at my non-www while I canonically preferred www — I had to reach out and ask them to update. Finally, if you use [JavaScript SEO](/javascript-seo/) frameworks that render client-side, the redirect from non-www to www must happen before JavaScript runs, otherwise search engines may see two pages. This is also relevant for [Core Web Vitals](/core-web-vitals/), because a redirect adds latency. Test your redirect chain with a tool to ensure it is a single 301.

I started my career assuming www was antiquated. I thought no www was always cleaner and more modern. That was wrong. Google has stated explicitly that it is not a ranking factor. I also let both versions resolve for months on a personal project because I was lazy. The consequence: I had two sets of search appearance results and a bunch of duplicate pages in the index. Fixing it required a 301 redirect, updating the [sitemap](/sitemap/), and resubmitting to Search Console. Another mistake: I relied on canonical tags alone, thinking that was sufficient. It is not. Google treats them as hints, not directives. Only 301 redirects consolidate signals completely. I also believed redirects lost link equity. In reality, modern search engines pass nearly all PageRank through a 301. I tested this on a site that had 200 backlinks to a page I redirected — within two weeks the new URL was ranking just as well. Finally, I neglected internal links. When I changed the preferred version, I only updated the big navigation links but left hundreds of relative paths unchanged. Because my CMS used relative URLs, they worked either way, but absolute URLs in content blocks pointed to the old version. That created a redirect chain on every page. The fix: a global search-and-replace. Use absolute URLs consistently in your templates.