What Exactly Counts as Technical SEO?
Ask five people what technical SEO means and you will get five answers, most of them about page speed. Speed is part of it, and one of the smaller parts.
The clearer definition is this: technical SEO is everything that determines whether a search engine can process your website properly.
This guide explains the fundamentals in plain terms — what each element does, how it fails, and how to check it yourself. For context on the wider discipline, see what SEO is and why it matters.
Key takeaways
- Technical SEO is about machine access and comprehension, not persuasion.
- Crawl access, indexation and rendering are the three foundations.
- One canonical version of every page prevents self-competition.
- Sitemaps aid discovery; internal links do the real work.
- Structured data states meaning explicitly in a machine-readable format.
- Search Console shows most technical problems for free.
What counts as technical SEO?
A useful test: if the change affects how a machine processes the site rather than what a human reads, it is technical.
- Server responses and status codes.
- robots.txt directives and crawl controls.
- Indexation signals such as noindex and canonical tags.
- Redirects and URL structure.
- Rendering of JavaScript-built content.
- XML sitemaps and discovery paths.
- Structured data in JSON-LD.
- Mobile usability and Core Web Vitals.
Titles, headings and copy are a different discipline. The overview of how the disciplines divide is in the guide to types of SEO.
How does crawl access work?
Search engines discover URLs through links and sitemaps, then request them. robots.txt tells them what they may request.
- Never block CSS or JavaScript — the page then cannot be rendered as users see it.
- Blocking a URL in robots.txt does not remove it from the index; noindex does that.
- Faceted navigation and search parameters can generate millions of low-value URLs.
- Links must be real anchors with href attributes to be followed.
Example. An ecommerce site blocked /assets/ to "save crawl budget". Google could no longer render product templates, and category rankings fell for two months until it was reverted.
What controls whether a page is indexed?
Three signals, plus a quality judgement you do not control directly.
| Signal | Effect | Common mistake |
|---|---|---|
| noindex | Excludes the page from the index | Left on live pages after a launch |
| Canonical tag | Names the preferred duplicate | Pointing every page at the homepage |
| robots.txt disallow | Prevents crawling, not indexing | Used to try to remove pages |
| Internal links | Signals importance and aids discovery | Orphaned key pages |
Beyond signals, Google decides whether a page is worth storing. Thin or near-duplicate pages are routinely crawled and left unindexed.
Why does rendering matter?
If content only exists after JavaScript runs, it depends on the search engine executing that code successfully.
- Check the rendered HTML, not the source, to see what is really available.
- Content behind clicks, hovers or infinite scroll is often unseen.
- Server-side rendering or static generation removes the risk entirely.
- Client-side routing must produce real URLs with server responses.
Google's URL Inspection tool shows the rendered output. It is the fastest way to settle an argument about whether content is visible.
How should URLs and redirects be handled?
URL discipline prevents a site from competing with itself, which is a more common problem than external competition.
- One hostname and one protocol, with everything else redirecting to it.
- Consistent trailing slash behaviour.
- 301 for permanent moves; 302 only for genuinely temporary ones.
- No redirect chains — point old URLs directly at the final destination.
- Readable, stable slugs; avoid changing them without cause.
After a migration, map every old URL to a new one before launch. Retrofitting redirects afterwards loses traffic that rarely returns in full.
What does structured data do?
It expresses facts about the page in a format that requires no interpretation: this is an organisation, this is its address, this is a service, this is an article by this author.
- Use JSON-LD rather than inline microdata.
- Mark up only what is visible on the page.
- Add BreadcrumbList on nested pages to communicate hierarchy.
- Validate markup, and re-validate after template changes.
It supports rich results and, increasingly, entity recognition in AI systems. It is not a ranking shortcut.
How much performance work is enough?
Enough that the site feels fast on a mid-range phone on mobile data. Chasing a perfect score has diminishing returns.
- Compress images and serve them at display size.
- Lazy-load anything below the fold.
- Audit third-party scripts; tag managers are frequent offenders.
- Reserve layout space for media to prevent shifting.
Which checks can you run yourself?
More than most people expect, and all of them free.
- Search Console page indexing report, filtered by page type.
- URL Inspection on one page per template, checking rendered content.
- A
site:search to spot duplicate or unexpected URLs. - Your own phone, on mobile data, for rendering and speed.
- Rich results validation on templates using structured data.
Work through the SEO website checklist for the wider picture, or read why technical SEO matters commercially for how these parts connect.
Frequently asked questions
Is technical SEO hard to learn?+
The concepts are straightforward; the diagnosis takes practice. Understanding crawling, indexing, canonicals and rendering covers most of what a business owner needs to supervise the work.
Does WordPress handle technical SEO automatically?+
Partly. Plugins manage sitemaps, canonicals and basic markup, but they also introduce duplicate archive pages, slow templates and conflicting settings that need review.
What is the difference between crawling and indexing?+
Crawling is requesting the page. Indexing is deciding to store and consider it. A page can be crawled repeatedly and never indexed, which Search Console reports explicitly.
Do I need an XML sitemap?+
It helps discovery, particularly on large or poorly linked sites, but it is not a substitute for internal links. List only canonical, indexable URLs in it.
How often should technical SEO be checked?+
Monthly for indexation and coverage, and immediately after any deployment, redesign or migration. Those events cause most new technical problems.
The short version
Technical SEO is about making your site legible to machines: reachable URLs, one canonical version of each page, content visible without interaction, and explicit markup.
Get those right once and the benefit persists. Ignore them and every pound spent on content is spent at a discount.
Technical foundations also decide whether AI answers can read you at all, which is where AI Overview optimisation picks up. If you want the current state of your site assessed, get in touch.
Related reading
Why is technical SEO important?
Technical faults quietly cap the return on every other marketing pound. The commercial case, with the failure patterns that cost most.
Read articleWhat is a technical SEO audit?
A real audit narrows hundreds of observations to the few problems costing you money. What to expect, and how to judge one.
Read articleSEO glossary
Definitions for the SEO terms you will meet in an audit, proposal or monthly report, grouped by subject.
Read article