The short version: Technical SEO is everything that makes your website easy for Google to find, crawl, understand and load fast, separate from the words on the page or the links pointing to it. If Google can’t crawl your site or your pages take eight seconds to load, brilliant content won’t save you. For most small business sites, fixing three or four things matters more than any 150-point audit an agency will try to sell you.
Technical SEO is the plumbing, not the paint
I explain it to clients this way: if your website is a house, content is the furniture and technical SEO is the plumbing and electrics. Nobody visits a house to admire the wiring, but if the wiring’s faulty, nothing else works. Technical SEO covers site speed, mobile usability, crawlability, indexing, secure connections (HTTPS), structured data, and how clean your site’s code and structure are behind the scenes.
It’s not about writing better copy or getting more backlinks. Those are different jobs (content SEO and off-page SEO). Technical SEO is purely about whether search engines can access, understand, and trust your site enough to show it to anyone in the first place.
The bits that make up “technical SEO”
When people say technical SEO, they usually mean some combination of these:
- Crawlability – can Googlebot get into your pages and follow the links between them? This is controlled by your robots.txt file and internal linking.
- Indexing – once Google crawls a page, does it decide to store it in search results at all? Not every crawled page gets indexed.
- Site speed – how fast pages load, measured through Core Web Vitals (Google’s specific speed and stability metrics).
- Mobile usability – whether the site works on a phone screen, since Google indexes the mobile version of your site first.
- HTTPS – whether your site has a secure certificate. Sites without it get flagged “Not Secure” in Chrome, which kills trust instantly.
- Structured data (schema markup) – hidden code that tells Google exactly what a piece of content is, like a recipe, a review, an event, or an FAQ.
- XML sitemaps and canonical tags – the map you hand Google of your site, and the way you tell it which version of a page is the “real” one when duplicates exist.
That’s the full list. Anyone stretching it beyond that into forty sub-categories is usually padding an invoice.
A real example: the 40,000-page problem
A few years back I worked with an online homeware retailer whose site had roughly 40,000 pages indexed in Google. The catalogue itself only had around 4,000 real products. The rest were duplicate URLs created by their filtering system, colour filters, price filters, sort-by-newest filters, each one generating a fresh URL that Google was dutifully crawling and indexing as a separate page.
Google was spending its “crawl budget” (the rough number of pages it’s willing to crawl on your site in a given period) chewing through 36,000 near-identical filter pages instead of finding and ranking the 4,000 pages that mattered. Their important product pages weren’t getting crawled often enough to pick up stock changes or new reviews.
The fix took about three weeks: we blocked the filter parameters in robots.txt, added canonical tags pointing every filtered variant back to the main category page, and resubmitted a clean sitemap with only the real 4,000 URLs. Four months later, organic traffic was up 22%, not because we’d written a single new word of content, just because Google could finally see the site.
How to check your own site’s technical SEO in under an hour
You don’t need to hire anyone to get a first read on this. Here’s the order I do it in:
- Step 1: Open Google Search Console (free) and check the “Pages” report under Indexing. Look at how many pages are indexed versus how many exist on your site. A huge gap in either direction is a red flag.
- Step 2: Run your homepage and two inner pages through Google’s PageSpeed Insights. Anything scoring under 50 on mobile needs attention.
- Step 3: Check the Core Web Vitals report in Search Console. This tells you if Google thinks your pages are “Good,” “Needs Improvement,” or “Poor” across real user data, not a lab test.
- Step 4: Download the free version of Screaming Frog (crawls up to 500 URLs free) and run it against your own site. Look for broken links, missing title tags, and duplicate meta descriptions in the output.
- Step 5: Type “site:yourdomain.com” into Google and scroll through what comes up. If you see pages you didn’t know existed, or pages that shouldn’t be public (staging pages, test pages, old campaign pages), that’s your indexing problem right there.
That hour will tell you more than most paid audits, because you’ll see your own site rather than reading a 40-page PDF summarising it.
The uncomfortable bit nobody in SEO likes to say
Here’s what most agencies won’t tell a small business owner with a 60-page website: you almost certainly don’t need a technical SEO audit at all. Crawl budget only becomes a real problem once you’re into the tens of thousands of pages. Structured data rarely moves the needle unless you’re in ecommerce, recipes, or local services with reviews. Most of the “150 technical issues found” reports I’ve seen for small sites are dominated by things Google has openly said don’t affect rankings, like minor image alt text gaps or slightly long URLs.
The people selling those audits know this. It’s easier to sell a scary 40-page report full of red flags than to say “your site’s fine, your problem is you’re only publishing once a month.” I’ve turned down work from small business owners because their actual issue was thin content and no consistency, not technical debt. Fixing the plumbing on a house with no furniture in it doesn’t sell the house.
Where technical SEO earns its keep is on large ecommerce sites, membership sites, or anywhere with thousands of URLs, faceted navigation, or a slow, ageing platform. If that’s you, it’s worth doing well. If you’re a local plumber with 30 pages, spend your money on getting reviews and writing better service pages instead.
Where images and audio fit into technical SEO
Two areas people forget entirely: images and audio/video content. If you’re using an AI image generator to create blog visuals or product mockups, those files still need compressing, proper file names, and alt text, otherwise you’re loading bloated PNGs that tank your Core Web Vitals score. I’ve seen a single unoptimised hero image add three full seconds to a page load.
Same logic applies if you’re publishing podcasts or video content: adding a written transcript through audio to text transcription work gives Google actual crawlable text to index from content it otherwise can’t “read,” and it’s one of the cheapest technical wins available to anyone doing regular video or audio content.
Who does this work
On most small teams, technical SEO gets split three ways: a developer fixes site speed and code-level issues, a marketer handles sitemaps and structured data, and increasingly, a well-briefed virtual assistant handles the ongoing monitoring, checking Search Console weekly, flagging new crawl errors, resubmitting sitemaps after big content pushes. You don’t need a specialist on payroll for most of this. You need someone checking a dashboard consistently, which is a task, not a skill you have to pay senior day rates for.
If you do decide to bring in outside tools rather than a person, worth knowing that pricing on the paid crawling and monitoring tools (Screaming Frog’s paid tier, Ahrefs, SEMrush) ranges from around £15 to £200 a month depending on how many URLs you’re tracking, which is covered in more detail in this breakdown of AI tool costs if budgeting is the sticking point.
The three things worth fixing first, in order
- 1. HTTPS and mobile usability. If either of these is broken, nothing else matters, Google won’t rank an insecure or unusable-on-mobile site competitively regardless of speed or schema.
- 2. Indexing accuracy. Make sure Google is indexing the pages you want, and not indexing the ones you don’t (staging pages, thin filter pages, old campaign landers).
- 3. Core Web Vitals. Once the above two are sorted, speed and stability are the highest-use fix left, because they affect every single page on the site at once.
Everything past those three is fine-tuning, useful on a big site, close to invisible on a small one.
Frequently asked questions
Is technical SEO the same as on-page SEO?
No. On-page SEO covers things within the content itself, like keywords, headings, and internal links within the text. Technical SEO covers the underlying site infrastructure, speed, crawlability, code and indexing, that sits behind the content regardless of what it says.
Do I need to hire a developer to fix technical SEO issues?
For some fixes, yes, particularly site speed and code-level issues on custom-built sites. For platforms like WordPress or Shopify, many technical fixes (sitemaps, canonical tags, basic image compression) can be handled through plugins or built-in settings without touching code.
How often should I check my site’s technical SEO health?
A quick Search Console check once a month is enough for most small business sites. Larger ecommerce or content sites with frequent new pages should check weekly, since indexing errors compound fast when you’re publishing regularly.
Can bad technical SEO stop my site ranking even with great content?
Yes. If Google can’t crawl or index a page, it doesn’t matter how good the writing is, that page effectively doesn’t exist in search results. Technical problems are the ceiling on everything else you do.
