MenuClose

Technical SEO: The Foundation for Google to Understand Your Website

Why loading times, indexing, structured data and a clean site structure decide whether search engines can find and understand a website.

By Ioan Serbanupdated on 7 min read

Technical SEO covers everything that makes sure search engines can find a website, read it and add it to their search index. That includes short loading times, clean code, working indexing and structured data.

Google describes this as a basic requirement. Without it, even the best content does not help, because Google does not capture the page properly.

When people think of search engine optimisation, they often think of keywords, blog posts or links first. These can matter, but they rely on clean technology. Without it, even good content goes to waste. As a bonus, technical SEO often makes a website more pleasant too: it loads faster and is easier to use.

The foundation for visibility

Before a page appears in the search results, Google has to find it, read it and add it to the search index. Finding and reading is called crawling. Adding it is called indexing.

If there are technical problems along the way, content appears only in part or not at all. That is why technical SEO is the foundation of every website, whatever the size of the business.

Short loading times and the Core Web Vitals

Visitors expect a page to be there within a few seconds. Google measures this with the Core Web Vitals, three values for the quality of a page:

  • LCP (Largest Contentful Paint): how fast the main content loads. Good is under 2.5 seconds.
  • INP (Interaction to Next Paint): how fast the page responds to a click. Good is under 200 milliseconds.
  • CLS (Cumulative Layout Shift): how stable the page stays while loading, without content jumping. Good is under 0.1.

That is harder than it sounds. According to worldwide data from the HTTP Archive (Web Almanac), only about half of all mobile websites reach all three values at once.

Slow pages also cost money. An analysis by Portent shows that just one extra second of loading time can cut the number of inquiries or purchases by around 7 %.

A page becomes fast through, for example:

  • optimised images in modern formats such as WebP,
  • smaller CSS and JavaScript files,
  • browser caching, which stores files on the visitor’s device for next time,
  • fast responses from the server,
  • loading content further down the page only when it is needed.

Fast pages please visitors and are easier for Google to capture.

Clean code structure

Search engines do not see the design; they read the HTML code. If it is cleanly built, they understand the content much more easily.

That includes:

  • exactly one main heading (H1),
  • logically ordered subheadings,
  • HTML elements with a clear meaning, for example for the header and the main content,
  • a meaningful page title,
  • a fitting description for the search results,
  • link texts that say where the link leads.

That way search engines recognise what a page is about and place it correctly.

The structure of the whole website matters too. Visitors and search engines should find information quickly.

A good structure has:

  • clear navigation,
  • clearly separated topics,
  • separate pages for individual services,
  • sensible links between the pages.

Internal links show search engines which pages belong together and which are especially important. Without them, some pages end up in a blind spot and are visited less often.

Controlling indexing

Search engines need to know which pages belong in the index and which do not. Google only spends a limited amount of time on each website, the so-called crawl budget. If that time goes on unimportant or duplicate pages, important pages are updated less often.

Important tools for this are:

  • the XML sitemap, a list of all pages for search engines,
  • the robots.txt, which tells search engines which areas to skip,
  • canonical tags, which name the main version of similar pages,
  • robots settings on individual pages, such as “do not index”,
  • correct status codes, the server’s answer on whether a page exists,
  • clean redirects from old addresses to new ones.

Mistakes here can mean that important content barely appears or not at all. Conflicting canonical tags are among the most common problems. Google then decides for itself which version to show. You can spot silent losses like these with a free website check.

JavaScript and rendering

Many modern websites build their content with JavaScript, for example with React, Vue or Angular. That is not bad in itself. But it becomes a trap when important text, links or structured data are only created in the browser.

Google reads a page in two steps. First it captures only the raw HTML, the page as the server sends it. Only later does it run the JavaScript and assemble the full page. This is called rendering. Until then, Google sometimes sees only an empty or half-finished page.

This matters even more for AI crawlers. A measurement by the hosting provider Vercel (December 2024) shows that the major AI crawlers, for example those for ChatGPT, Perplexity or Claude, do not run JavaScript. They read only the raw HTML. Title, main heading, main text and navigation should therefore already be in the first HTML.

Two approaches are more robust here than content built in the browser: server-side rendering (SSR), where the server delivers each page finished, and static pages (SSG), which are created as finished HTML files when the site is published. This website uses static pages too, and why I rebuilt it is explained in a separate article. Which technology suits your website is covered in the comparison WordPress or a custom website.

Structured data

Structured data are extra details in the code of a page. They tell search engines clearly what the page is about, for example:

  • who the business is,
  • which services it offers,
  • where it is located,
  • how to reach it,
  • when it is open.

Structured data do not bring better rankings automatically. But they make the content easier to understand and raise the chance of an enhanced display in the search results. It is important that every page carries all the details it refers to itself. And they should be in the first HTML, not added later with JavaScript. Google can usually handle both, but AI crawlers only the first.

Optimised for phones

More than half of all visits on the web now come from smartphones. Google judges websites by their phone version. So the mobile view decides your visibility directly, however good the page looks on a computer.

That includes:

  • a layout that adapts to every screen size (responsive design),
  • easy-to-read font sizes,
  • buttons that are big enough,
  • no sideways scrolling,
  • short loading times on mobile networks too.

HTTPS and security

Security is part of technical SEO as well. A website should only be reachable over HTTPS, so that data is sent encrypted.

Regular updates, a securely configured server and clean redirects make the foundation stable as well.

Technical SEO and ongoing SEO

Technical SEO makes sure that search engines can process a website well. It is the foundation, but only one part of search engine optimisation.

Good rankings do not come from clean technology alone. Just as important is good content that matches what customers search for, and a website that keeps developing. That is why SEO is usually a long road: businesses add content, update pages and watch how their visibility develops. Google also changes its rules all the time.

The technology, on the other hand, is usually checked at longer intervals. It creates the conditions for every other step to work. What those steps are is shown in the overview of the most important SEO factors. For businesses that want to be found locally, the article on local SEO explains the next steps.

With every business website I build, this technical foundation is part of the fixed price: page titles, structured data, a sitemap and short loading times.