A technical SEO checklist for custom-built websites
The technical foundations we build into a custom website from the first release, so that search engines can crawl, understand and index it.
· 4 min read · By WebSurge Technologies
Search engines can only rank what they can crawl and understand. A custom website gives you full control of the technical details, and that control cuts both ways: nothing is added for you. This checklist covers what we build in from the first release.
1. One address for every page
- Choose one canonical host (for example www) and one URL style (for example a trailing slash), and redirect everything else to it with a single 301.
- Give every page a
rel="canonical"link that matches its real address. - Force HTTPS and avoid redirect chains.
2. Titles, descriptions and headings
- Every page has a unique title and meta description that describe the page honestly.
- Each page has exactly one
h1, and headings go in order. - Images have useful alt text, and decorative images have empty alt text.
3. Crawl control
- Publish a
sitemap.xmlthat lists only canonical, indexable pages that return status 200. - Publish a
robots.txtthat points to the sitemap and blocks only what should not be crawled, such as an API directory. - Do not put
noindexpages in the sitemap, and do not block pages with robots.txt if they need anoindexto be seen. - Return a real 404 status for missing pages, with a helpful page for visitors.
4. Structured data
Structured data in JSON-LD helps search engines understand who you are and what a page is. A typical business site uses Organization, WebSite, WebPage, BreadcrumbList and, where they fit, Service, Article and FAQPage. Mark up only what is visible on the page, keep it consistent with the visible text and validate it with Google’s Rich Results Test and the Schema Markup Validator.
5. Performance and mobile
- Meet the Core Web Vitals thresholds, as described in Core Web Vitals budgets for custom websites.
- Use a responsive layout and a viewport meta tag, and keep tap targets large enough.
6. Social sharing and AI readers
- Add Open Graph and Twitter Card tags with a 1200 by 630 pixel image, so that shared links look right.
- Keep the content readable as plain HTML. Pages that render their text only in JavaScript are harder for crawlers and assistants to read.
- A
llms.txtfile that summarizes the site for AI assistants is a community proposal, not a standard. It costs little to publish, but no search engine requires it.
7. Check in Search Console
After launch, submit the sitemap in Google Search Console, inspect a few key URLs, and watch the Pages, Core Web Vitals and Enhancements reports. Fix the cause of a warning rather than hiding it.
Where we have used this
The LuxPines build runs a check command over the finished site that verifies broken links, images, metadata, headings and structured data before anything is uploaded. We apply the same discipline to this website.
Internal links and site structure
- Link every important page from at least one other page, and keep key pages within a few clicks of the home page.
- Use hub pages. A services page that links to each service, and a case-study page that links to each case, helps visitors and crawlers see how pages relate.
- Write descriptive link text. “Read the case study on LuxPines” says more than “click here”.
- Add breadcrumbs, and mark them up with structured data.
Content and trust signals
Google’s quality guidelines talk about experience, expertise, authoritativeness and trust. In practice, make it easy to see who is behind the site: an about page, a contact page, a privacy policy and terms. Show dates on articles, explain where numbers come from and do not claim results you cannot support. Original, specific content, written from real work, tends to serve both visitors and search engines better than generic text.
Rendering and JavaScript
Serve the main content as HTML. Search engines can run JavaScript, but it is slower and less reliable, and many assistants and social previews do not run it at all. A site that is complete without JavaScript, and improved by it, is the safest design.
Common Search Console messages and what they usually mean
- Duplicate without user-selected canonical. Several URLs show the same content and none says which is preferred. Add a canonical link, and redirect the alternatives.
- Page with redirect. The URL redirects, so it is not indexed. This is normal for old addresses. Make sure that internal links point to the final address.
- Not found (404). A listed or linked URL does not exist. Fix the link or redirect it if the page has moved.
- Soft 404. A page that looks empty or like an error returns status 200. Return a real 404 or add real content.
- Submitted URL marked “noindex”. A page in the sitemap tells Google not to index it. Remove it from the sitemap or remove the tag.
- Crawled, currently not indexed. Google saw the page and chose not to index it yet. Improve its uniqueness and internal links, and give it time.
Treat each message as a question about one page. Fix the cause, then use the validate option in Search Console.
Keep reading
- Case study: LuxPines
- Core Web Vitals budgets for custom websites
- Our cloud, performance and architecture service
Have a project in mind? Book a free consultation.