Technical SEO: a checklist so Google can crawl, understand and index your website

Technical SEO is the least visible part of search optimisation and the one that holds you back most when it fails. You can have the best content in your industry, but if Google can't crawl it, doesn't understand it or decides not to index it, it won't appear in the results.

This checklist summarises the checks we carry out when starting an SEO project. You don't need to be a developer to review most of them; you do need access to Google Search Console.

1. Crawling: can Google reach your pages?

• robots.txt. Check it doesn't block important sections by mistake. It's common to find websites that inherited a full block from the staging environment.
• Crawlable internal links. Menus and links should be real HTML links. If navigation depends on complex interactions, Google may not discover some pages.
• Orphan pages. Every important page should receive at least one link from another page on the website.
• Server errors. Check the crawl stats report and 5xx errors in Search Console.

2. Indexing: which pages are on Google?

In Search Console, the page indexing report shows which URLs are indexed and why others aren't. Pay attention to:

• Important pages marked “Crawled – currently not indexed”, which usually indicates content that is thin or poorly differentiated.
• Pages excluded by a noindex tag that shouldn't be there.
• Duplicate pages without a clear canonical.

3. Canonicals and duplicate content

The canonical tag tells Google which is the main version of a page. Check that:

• Each indexable page has a canonical pointing to itself (or to the main version if it's a duplicate).
• Versions with and without a trailing slash, with and without “www” or with campaign parameters don't coexist without a canonical that unifies them.
• Domain redirects (http to https, with or without www) are permanent and direct, without chains.

4. XML sitemap

The sitemap helps Google discover your important pages. It should:

• Include only indexable URLs that return a 200 status code.
• Exclude noindex pages, redirects and errors.
• Update automatically when you publish new content.
• Be submitted in Search Console and referenced in robots.txt.

5. Architecture and internal linking

A good architecture lets you reach any important page in a few clicks and tells Google which pages are priorities.

• Group content by topic and link related pages to each other.
• Link from blog articles to service pages when it's relevant for the reader.
• Use descriptive link text instead of “click here”.
• Add breadcrumbs to reinforce the hierarchy.

6. One page per search intent

When several pages try to rank for the same search, they compete with each other. This is known as cannibalisation. To avoid it, give each page a distinct main intent and document that allocation in a keyword map.

7. Performance and Core Web Vitals

Page experience influences rankings and, above all, conversion. Check the Core Web Vitals report in Search Console, based on real-user data:

• LCP (loading of the main content): optimise images, fonts and server response time.
• INP (responsiveness to interaction): reduce unnecessary JavaScript and third-party scripts.
• CLS (visual stability): reserve space for images, ads and late-loading elements.

8. Mobile and accessibility

Google mainly uses the mobile version of your website for indexing. Check that content, links and structured data are the same on mobile and desktop. An accessible website — with good contrast, alt text on images and a logical heading structure — is also usually easier for search engines to understand.

9. Structured data

Schema.org markup helps Google interpret content: organisation, services, articles, breadcrumbs, FAQs… Two basic rules:

• Markup must describe content visible on the page.
• Don't mark up ratings or reviews that don't exist or that you can't verify.

Validate the markup with Google's Rich Results Test and remember that having it doesn't guarantee a rich result will be shown.

10. Status codes and errors

• Pages that don't exist should return a real 404, with a useful page that helps people keep browsing.
• Removed content that had traffic or links should be redirected (301) to the closest equivalent page.
• Avoid “soft 404s”: empty or error pages that return a 200 code.

How to prioritise

You don't have to fix everything at once. Start with what blocks (crawling and indexing), continue with what confuses (duplicates, cannibalisation, canonicals) and finish with what improves (performance, structured data, linking).

If you'd rather we reviewed it for you, we can carry out a technical audit as part of our SEO service.

Last updated: 26 October 2026. Questions about this topic? Write to us.