Drupal SEO Checklist

Start with the basics that too many sites skip. Enable clean URLs – Drupal 8+ does this by default, but check your settings.php to confirm. Then install Pathauto and set up patterns for each content type so a blog post at /node/42 becomes /blog/how-to-optimise-drupal. Without this, you leave crawl efficiency on the table. Next, force HTTPS sitewide and configure the HSTS header. I once inherited a site that still served HTTP for half its pages; the duplicate content issue took me a day to untangle. Use the Redirect module to map HTTP to HTTPS with a 301.

Canonical URLs are non-negotiable. Drupal’s core sets them for nodes, but views and taxonomy pages often miss one. Install the Metatag module and enable canonical tags for all entity types. Then generate an XML sitemap using the Simple XML Sitemap module – it handles language variants, which is critical if you run a multilingual site with [hreflang tags](/hreflang-tags/). Submit the sitemap to Google Search Console and watch the indexed pages count.

Now review robots.txt. Drupal’s default allows everything on standard paths, but that includes /node/add and /user/register. Block those. For taxonomy term pages with thin content, consider adding a noindex directive via the Metatag module. I also set up [301 redirects](/301-redirect/) for any old URLs that survived a site migration – you only need one broken link to lose a ranking.

Structured data comes next. The Schema.org Metatag module lets you add JSON-LD markup per content type. For an article page, I add Article schema with headline, image, and datePublished. For events, add Event schema. It doesn’t improve rankings directly, but it gets you into rich results. Finally, run a quick [SEO audit](/seo-audits/) to catch missing meta descriptions, thin H1s, and slow pages. Fix the H1 hierarchy first – I always check that each page has exactly one H1 and that headings follow a logical order.

There’s no such thing as a perfect Drupal SEO setup, and chasing every possible improvement will waste time that should go into content. I stop when the following conditions are met.

First, the crawl fundamentals are solid: clean URLs, HTTPS, canonical tags, a working sitemap, and robots.txt that doesn’t block important paths. If I still see pages in the index that shouldn’t be there, I check the canonical tags and redirects before touching any module. Second, I stop after I’ve verified structured data on three representative pages – the home page, a typical article, and a product page. I use Google’s Rich Results Test, not the Schema validator, because the validator flags warnings that don’t affect search. Third, if the site has fewer than 50 pages, I don’t spend more than a few hours on Drupal-specific tuning. Core settings plus Pathauto and Metatag are enough.

The real stop condition is diminishing returns. If the site already loads in under 2.5 seconds (on mobile) and has no duplicate content issues, I move on to [website structure](/website-structure/) and internal linking. The last thing I’d do is enable caching modules like Internal Dynamic Page Cache – that’s for the infrastructure team. My rule: if a checklist item doesn’t change how Googlebot sees a page, skip it.

Honestly, for my first few Drupal sites, I installed Pathauto and Metatag and called it done. I assumed the sitemap module would exclude low-quality taxonomy terms automatically – it doesn’t. I wasted three months wondering why dozens of /taxonomy/term/X pages were indexed and competing with actual content. The fix was adding a noindex override in Metatag for all taxonomy term pages with fewer than two child nodes. I should have checked the site: query in Search Console from day one.

Another mistake: I relied too heavily on the core Redirect module for URL aliases without testing the redirect chain. Pathauto changes the alias when you update the title – suddenly old URLs 301 to a new alias, but if that new alias changes again, you get a 302 loop. I now set the ‘Update action’ in Pathauto to ‘Do nothing’ and manually add redirects when I rename content. It sounds slower, but it has saved me from hundreds of 404s in the logs.

I also used to ignore the robots.txt for multilingual setups. Drupal’s default allows paths like /{lang}/search/ and /{lang}/node/. Blocking those earlier would have saved crawl budget for the actual localised pages. And I spent too long tweaking the max_url_length setting for the sitemap – the default 2048 characters is fine. Learn from my time sink.