What is technical SEO?

Crawling: can Google reach every page

Search engines discover pages by following links, starting from URLs they already know about and hopping outward from there. Googlebot works through a crawl queue, and if a page has no internal links pointing to it, sits behind a login wall, or is blocked by a Disallow rule in robots.txt, it can effectively be invisible to Google no matter how good the content is. This matters more than most site owners expect, because search engines don't know your site exists just because it's live: they need a path to every page, and that path has to survive redesigns, replatforming and CMS migrations, which is exactly when a stray robots.txt rule most often gets left behind. Two checks catch most of the damage early: confirm your XML sitemap actually lists every page you want found, and read robots.txt line by line rather than skimming it, since a single wildcard rule written for a staging environment can quietly disallow an entire section of the live site.

Indexing: found is not the same as ranked

Crawling and indexing are two separate decisions Google makes, and conflating them is one of the most common technical SEO mistakes. A page can be crawled successfully and still get excluded from the index, usually because Google judges the content too thin to be worth storing, treats it as a near-duplicate of another page on your site, or finds a noindex meta tag that was meant for a staging server and never got removed before launch. None of this is visible from browsing your own site, since the page loads fine in a browser either way; it only shows up in Search Console's Page Indexing report, which lists every excluded URL against the specific reason Google gave for leaving it out. A site with 200 pages and only 60 indexed has a real, fixable problem, and it's worth chasing down before spending any time rewriting content, because no amount of better copy fixes a page Google has already decided not to store.

Site structure and internal links

How your pages link to each other determines how easily both crawlers and visitors move through your site, and it also determines how the authority your site has earned gets distributed. A flat structure, where any important page sits no more than two or three clicks from the homepage, spreads that authority evenly and gives Google a clear signal about which pages matter. The two structural faults that show up most often are orphan pages, which have no internal links pointing to them at all and can go unindexed indefinitely even when they're technically crawlable, and categories or product pages buried five or six clicks deep, which get crawled less often because Google allocates less attention to pages it has to work hard to reach. Internal links do double duty: they're navigation for people, but they're also a vote of confidence for search engines, so the pages you link to most from your best-performing pages are the ones Google will treat as most important.

Structured data (schema markup)

Schema markup is code, usually written as JSON-LD, that states facts about a page in a format machines can parse without inferring them from the surrounding text: this is a product, this is its price and availability, this is an FAQ and here are the exact question-and-answer pairs. Without it, Google has to guess at that structure from the visible page, and guessing is exactly where ambiguity creeps in. Schema doesn't guarantee a rich result like a star rating or an FAQ dropdown in search results, but skipping it removes any chance of one, which matters because rich results tend to take up more space and draw more attention on the results page. Organization, Product, FAQPage and BreadcrumbList are the schema types most sites are missing, and they're also the ones with the clearest mechanical link between having them present and being eligible for the search feature they unlock.

Speed and mobile usability

Core Web Vitals and mobile usability sit inside technical SEO because Google measures them directly as ranking signals, rather than inferring them the way it infers relevance from your copy. Core Web Vitals cover loading speed, interaction responsiveness and visual stability, and a page that's slow to load or shifts around while someone's trying to tap a button fails these measurements regardless of how well-written the page is. Mobile usability works the same way: if text is too small to read, buttons sit too close together to tap accurately, or the layout breaks on a phone screen, that's a technical fault Google can detect, not a content gap. The reason this belongs in the same conversation as crawling and indexing is that it's the same category of problem: something happening at the infrastructure and code level that no amount of editing the words on the page will ever fix.

Canonicalisation and duplicate content

Duplicate content is one of the quieter technical SEO problems, and it's worth separating from the indexing issue above because the mechanism is different. Most sites don't set out to duplicate content, but ecommerce platforms generate near-identical URLs for the same product through size and colour filters, content management systems often serve the same page at both a trailing-slash and non-trailing-slash URL, and www versus non-www versions of a domain can both resolve without a redirect in place. Left alone, Google has to guess which version is the real one to rank, and it doesn't always guess the version you'd choose. A canonical tag tells Google directly which URL is authoritative, consolidating any ranking signal spread across the duplicates onto a single page instead of splitting it thin across several. Checking for missing or conflicting canonical tags, and for www/non-www and trailing-slash redirects being set up correctly, catches a class of problem that content edits alone can never touch.

How to check yours without guessing

Most of what's covered above is invisible until you specifically go looking for it: crawl errors, index exclusions, missing schema and Core Web Vitals failures rarely surface just from browsing your own site the way a visitor would. That's the core difficulty with technical SEO as a category - it's not that the fixes are hard, it's that the problems don't announce themselves. A scan that checks crawlability, indexing signals, structured data, canonicalisation and Core Web Vitals together, in one pass, is a faster way to find out which of these is actually costing you than working through each one manually with a different tool. AuditHQ's Technical suite runs hundreds of these checks in one pass, which is the fastest route from 'something feels off with our search visibility' to a concrete, prioritised list of what to fix first.

Frequently asked questions