When people hear SEO, they usually think of keywords and content. But whether a search engine can reach your website, find its pages, read them and understand them correctly depends on the technical foundation. If that foundation is weak, even well-written pages may never show up in search results. The 12 checks below can be used before launching a new website or when auditing an existing one. They do not replace a content strategy, but they give it the ground it needs to work.
Accessibility and indexing
1. HTTPS and a single address
The website should open only over HTTPS, and the HTTP version should send visitors to HTTPS with a permanent (301) redirect. The same applies to the "www" and non-"www" versions: choose one as the main address and redirect the other to it. Otherwise the search engine sees several copies of the same page.
2. robots.txt
This file tells search robots which parts of the website they should not crawl. The most common mistake is a rule left over from the staging environment that blocks the entire site. The file should block the admin panel and technical sections while leaving public pages open. It can also point to the address of the sitemap.
3. sitemap.xml
A sitemap gives search engines the list of pages you want indexed. It should update automatically, contain only public pages that actually exist, and leave out addresses that redirect, have been deleted or are excluded from indexing. On large websites the sitemap is split into several files, with an index file tying them together.
4. Indexing control
Pages that should not appear in search (internal search results, the cart, user accounts, thank-you pages) should be excluded from indexing with the robots meta tag. Conversely, check that no important page has been blocked with "noindex" by accident. The indexing reports in search engines' webmaster tools are useful for this.
Structure and duplicate content
5. URL structure
Addresses should be short, readable and stable: lowercase letters, hyphens between words, no meaningless parameters. Once an address is published, it should not change; if it has to, set up a 301 redirect from the old address to the new one.
6. Canonical
The same content can open at several addresses: filter and sorting parameters, campaign tags, pagination. The canonical tag tells the search engine which address is the main one. Every indexable page should have a correct canonical pointing either to itself or to the main version.
7. Redirects and error pages
A page that does not exist should return a real 404 status code, not a 200 code with "not found" text. Redirect chains (A to B, B to C) should be shortened, and internal links should point directly to the final address. When moving from an old website to a new one, map every important old address and redirect each to the matching new page: this is the step most often forgotten when a website is rebuilt.
Failing to redirect old addresses during a rebuild is the fastest way to lose search rankings and the value of external links built up over years.
Speed and user experience
8. Loading speed and Core Web Vitals
Core Web Vitals are metrics that measure how quickly a page becomes visible to the user, how quickly it responds, and whether elements jump around while it loads. Typical steps to improve them:
- serve images at the right size and in modern formats, and lazy-load those below the fold;
- reserve space for images and ad blocks in advance so the page does not shift;
- cut unnecessary JavaScript and third-party scripts;
- use server caching and browser caching for static files.
9. Mobile-friendliness
Search engines evaluate a website mainly by its mobile version. Text should be readable without zooming, buttons should be easy to tap with a finger, there should be no horizontal scrolling, and the mobile version should not hide important content that the desktop version shows.
Technical markup of content
10. Headings and metadata
Every page should have a unique title and meta description. A page should have one main heading (h1), with subheadings following a logical hierarchy. Meaningful alt text on images matters both for accessibility and for image search.
11. Multiple languages and hreflang
If the website is in several languages, each language version should have its own address (for example, with a language prefix) and be linked to the others with hreflang annotations. These annotations must be reciprocal: each version points to the others and to itself. Switching the language automatically based on the user's browser, or showing all languages at one address, prevents search engines from seeing the language versions correctly.
12. Structured data
Data marked up with the Schema.org vocabulary tells the search engine precisely what a page is about: an organisation, an article, a product, questions and answers, a breadcrumb trail. The markup should describe only content that is actually visible on the page and should be tested with search engines' validation tools.
Common mistakes
- Leaving the staging environment open. A test version of the website stays open to indexing at a separate address and creates duplicate content with the main site. The test environment should be closed with a password or an indexing ban.
- Keeping the ban at launch. The opposite happens too: a "noindex" or robots.txt block added during testing is forgotten on the new website.
- The same title on every page. An identical title inherited from a template gives the search engine no way to tell pages apart.
- Content loaded only by scripts. The main text and links should be in the page's initial HTML so robots can see them reliably.
- Heavy images. Large, unoptimised images are the simplest and yet most common cause of slow loading.
- Broken internal links. Links to pages that have been deleted or moved send both users and robots nowhere.
How to organise the check
You do not always need sophisticated tools. Search engines' free webmaster tools give you the essentials on indexing, sitemaps, mobile-friendliness and speed, while the browser's developer tools let you check status codes, canonical and meta tags. On large websites, crawlers that walk through every page quickly find broken links, duplicate titles and redirect chains.
This list is not a one-off task. A practical routine looks like this:
- Before launch, all 12 points are checked in the staging environment, especially the robots.txt and noindex settings.
- Right after launch, the website is added to search engines' webmaster tools, the sitemap is submitted and the redirects from old addresses are checked.
- On a regular basis, indexing errors, speed metrics and broken links are monitored.
- After every major change (a new section, a redesign, a platform change), the check is repeated.
Conclusion
Technical SEO makes sure your website is open, clear and fast for search engines: correct addresses, indexing control, speed, mobile-friendliness and multilingual setup. Once this foundation is in place, content and marketing efforts deliver better results. The most reliable approach is to build these requirements in while the website is being developed rather than fixing them afterwards. If you would like to check your website's technical health or build a new website that is ready for search, get in touch with TRUE.
