What technical SEO actually is
Technical SEO is everything that has to be true about your website before a single word of on-page work can earn a ranking. The page has to be crawlable. The HTML has to render. The content has to be indexed. The signals have to reach Google in the form Google expects. Get any of those wrong and the rest of your SEO budget is being spent on a page that nobody at Google has ever seen.
Where on-page is what an editor touches and off-page is what other sites do for you, technical is what your developer (or you, with developer access) does to the site itself. Robots.txt rules. Canonical tags. Server response codes. XML sitemap generation. Redirect chains. JavaScript rendering. Hreflang annotations. The lot.
The classic textbook definition lists about eight things. The honest 2026 definition is broader because Google now does more work to understand pages, and there are more ways for a page to fail before it ever reaches a SERP. That includes Core Web Vitals (now a baseline expectation, not a tiebreaker), JavaScript rendering (most modern frameworks need work to be SEO-safe), hreflang for any site serving multiple countries, and structured data that AI summaries lean on.
Two useful distinctions to start with.
First, crawlable is not the same as indexable, and indexable is not the same as rankable. A page can be crawled but blocked from indexing. A page can be indexed but never rank because the technical signals say it is a duplicate or low priority. A page can rank but lose impressions because the title in the SERP is rewritten badly. Each stage has its own failure modes and its own diagnostics. The pillar walks through all four stages.
Second, technical SEO splits into hygiene and architecture. Hygiene is the work you do once and check quarterly: HTTPS, robots.txt, a clean sitemap, no broken redirects, no accidental noindex tags. Architecture is the structural work that takes thinking: URL patterns across thousands of pages, internationalisation, JavaScript rendering strategy, log file analysis on a six-figure-URL site. Most Perth small businesses only need hygiene. E-commerce platforms with 30,000 SKUs need both, with the architecture work dominating.
One last framing. Technical SEO is the layer that disappears when it is working. Nobody notices a clean sitemap. Everybody notices when a botched redirect kills 40 percent of organic traffic over a long weekend. The job pays in catastrophes avoided, not in glory. That is part of why it is the half of SEO most often ignored until something breaks.
Why technical SEO matters for Australian businesses
Two reasons. The first is defensive: getting it wrong costs you everything. The second is offensive: getting it right opens a moat.
On the defensive side, every Perth agency has a horror story about a client whose rankings vanished because a developer pushed a staging robots.txt to production. We have inherited at least a dozen sites where a single misconfigured canonical tag was diverting authority from the right page to the wrong one. Australian e-commerce sites with faceted navigation (filters for size, colour, brand) routinely waste 80 percent of their crawl budget on URLs nobody should ever land on. Each one is a multi-week recovery if you catch it. A multi-quarter recovery if you do not.
On the offensive side, technical SEO is one of the few areas in SEO where most of your competitors have not done the work. Title tags get rewritten because they sit in the page builder. Meta descriptions get filled in because the WordPress plugin nags. Robots.txt sits there from the day the developer set up the site, untouched for years. Run a crawl over the top ten ranking sites for almost any Australian commercial query and you find broken hreflang, missing canonical tags, render-blocking JavaScript and orphan pages on most of them. Cleaning that up on your own site is a defensible edge.
There is also a category effect specific to the Australian market. We serve a few patterns that magnify the technical SEO payoff:
- Multi-suburb service businesses. Perth tradies, healthcare and professional services routinely build 20 to 60 suburb pages. Without proper canonical strategy and internal linking, those pages compete with each other and none of them rank. With the technical work done, they cover the metro area and the calls come in.
- Large mining-services catalogues. WA mining-services sites and equipment-supply businesses sometimes carry thousands of product SKUs. The crawl budget, faceted navigation and canonical work decide whether Google even tries to index them.
- Retail and e-commerce with seasonal variants. Different SKUs, different sizes, different states. Hreflang for AU vs NZ. Canonical patterns for stock-vs-sold. The technical work is what stops Google merging your product listings into a soup of duplicates.
- Professional services with location duplication. Legal, accounting, healthcare practices with offices in multiple WA towns. Each location needs to be canonicalised, structured-data'd and internally linked correctly, or Google folds them all into one entry and the other locations vanish.
The AI search shift makes the technical work more valuable, not less. AI Overviews, ChatGPT Search and Perplexity all sit on top of indexed content. If Google never indexed your page, the AI engines do not see it either. The same technical hygiene that wins the ten blue links also wins the AI citation. Skip it and you become doubly invisible. Best industry estimates put AI Overview click depression at 20 to 30 percent on informational queries, with commercial queries far less affected. Either way, the pages that get cited inside the AI summary tend to be the same pages that already ranked. Technical SEO is what keeps you in that pool.
One last point on the maths. Technical wins compound differently from content wins. A new blog post earns traffic for the keywords it targets. A technical fix earns traffic across every page on the site. We have seen single redirects unblock thousands of indexed pages overnight. We have seen one canonical update lift an entire category by ten positions. The cheapest unit of organic traffic you will ever buy is the morning your developer spends fixing the eight things in our standard technical audit.
The 12 sub-topics that make up technical SEO
This pillar splits into 12 chapters. Each one stands alone, but they are listed here in the rough order we work through them on a new client. Audit first. Then crawl and indexing. Then the structural pieces (sitemaps, redirects, canonical tags). Then the advanced topics. Site migrations are the final boss; we save that for the end because everything in the pillar feeds into doing one safely.
- The technical SEO audit, step by step. Our actual audit process. The 28 checks we run, in the order we run them, on every new client site.
- Crawl budget explained. What it really means, when it matters (and when it does not), and how to stop Googlebot wasting it on URLs you do not want ranked.
- Canonical tags explained. The single most-misunderstood tag in SEO. What it does, when to use it, the three patterns that work and the four that quietly destroy rankings.
- Robots.txt and meta robots. Two related tools that do very different jobs. How to use each one without locking Google out of your own site.
- XML sitemaps explained. What goes in, what stays out, how big is too big, and the common WordPress plugin defaults that hurt more than they help.
- 301 vs 302 redirects. Permanent vs temporary, what each one signals to Google, the chain limits, and the redirect mistakes that bleed traffic the slowest.
- Hreflang for international SEO. If you serve more than one country or language, this tag exists to stop Google blending the variants together. The honest version of how to set it up.
- JavaScript SEO essentials. React, Vue, Next.js, Nuxt and the rest. What renders, what does not, the three rendering strategies and when each one is the right call.
- HTTPS and SSL for SEO. Why it still matters in 2026, how to migrate without losing rankings, and the cipher and certificate gotchas that quietly trip people up.
- Mobile-first indexing. Google now indexes the mobile version of your site as the canonical version. What that means in practice if your desktop and mobile templates differ.
- Log file analysis for SEO. The advanced diagnostic for large sites. How to pull Googlebot's actual behaviour out of your server logs and use it to spot crawl waste.
- Site migration SEO checklist. The 70-step checklist we follow on every replatform, redesign, domain change or HTTPS migration. The reason most migrations bleed traffic is that nobody had a checklist.
Our framework: the crawl-render-index-rank pipeline
The 12 topics above sit inside a four-stage pipeline that every page on your site passes through. We use this framework on every audit because it tells you, in seconds, where a page is failing. Skip a stage and the page never makes it to the next one.
The four stages, in order:
Stage 1: Crawl
Does Googlebot even reach the page? This stage is about discovery and access. The relevant tooling is robots.txt, the XML sitemap, internal linking density (every orphan page is a crawl problem), crawl budget on large sites, and server response codes. If a page is blocked, never linked, or returns a 5xx error, it never reaches stage two.
Stage 2: Render
Can Googlebot turn the HTML and JavaScript into the page a user would see? This is where modern stacks fall over. A React or Vue app that builds the page client-side might serve Googlebot an empty <div id="root">. JavaScript SEO is the chapter that covers this. Server-side rendering, static generation, dynamic rendering and hybrid approaches all have their place. Pick wrong and your content is invisible.
Stage 3: Index
Does Google decide the page is worth keeping in the index? This is where canonical tags, meta robots (noindex), duplicate content checks and the sitemap converge. A page can be perfectly crawled and rendered, but if the canonical points elsewhere or the meta robots says noindex, it never lands in the index. Indexing failures are the single most common reason a new page does not rank.
Stage 4: Rank
Now we are into territory where on-page, off-page and technical all converge. The technical contributions at this stage are page speed (Core Web Vitals), mobile-first rendering, structured data that earns rich results, and HTTPS as a baseline trust signal. The big content and authority levers sit in on-page SEO and off-page SEO, but the technical baseline has to be there or the on-page work earns nothing.
Why the pipeline framing is useful
When a page is not getting traffic, we ask: which stage is it failing at? The diagnostics for each stage are completely different. Crawl failures show up in Search Console's Crawl Stats and server logs. Render failures show up in Inspect URL's "Tested page" tab inside Search Console. Indexing failures show up in the Coverage report. Ranking failures show up in the Performance report. Map the symptom to the stage and the fix list shrinks by 90 percent.
We get a lot of calls that go "my page is not ranking, can you help?". The honest answer is always "let me check which stage it is failing at". Two out of three times, the page is not even indexed. The owner has been writing more content for six months when a single canonical tag fix would have unblocked the lot.
Where most businesses get stuck
Same patterns, same Perth sites, every audit. Here are the ones we find most often:
- A weekly five-minute check of Search Console's Indexing report. Any spike in "Not indexed" pages gets investigated immediately.
- The XML sitemap is generated dynamically and contains only URLs you want indexed: no draft pages, no parameter URLs, no thank-you pages.
- The robots.txt has been read by a human at least once in the last 12 months.
- Every redirect is a 301 unless there is a documented reason it is not. A spreadsheet of every redirect lives somewhere reachable.
- Page templates are mobile-first by design, not desktop-first with mobile responsiveness bolted on.
- Site migrations have a checklist (ours has 70 items). Nothing ships without it.
- The robots.txt was set up by the developer in 2019 and has never been touched. Half the team does not know it exists.
- The XML sitemap is generated by a plugin that includes every URL the CMS knows about: drafts, archives, tag pages, paginated comments, the lot.
- Canonical tags self-reference on every page by default. Nobody has checked whether the parameter pages, faceted pages and pagination pages need different canonicals.
- Redirects are chained. The 2019 redirect chains to the 2021 redirect chains to the 2024 redirect. Three hops, lost authority, slow page.
- HTTPS migrated five years ago but a handful of internal links still point to
http://. Every one of those is a 301 hop the page does not need. - The site runs on a JavaScript framework but nobody knows whether Googlebot is rendering the pages or seeing the loading shell.
- Site migration was done over a weekend by a developer who had never done one. Three months later, traffic is down 60 percent and nobody knows why.
If you do nothing else this month, do these three: read your robots.txt with fresh eyes, check Coverage in Search Console for any spike in excluded pages, and crawl your top 100 URLs with Screaming Frog looking for noindex tags. Twenty minutes for the lot. We find a billable problem in roughly half the audits we run on this checklist alone.
The Perth and WA pattern we keep seeing
We sit down with a new technical client every other week. The conversation has a shape to it.
A Perth trades business with 18 suburb pages, all built off the same template, every single one canonicalised to the homepage. The owner cannot understand why the suburb pages do not rank. The canonical is telling Google those pages do not deserve to exist as their own URL. One change, 18 pages, two months later they own the local pack across the metro area.
A Fremantle e-commerce site running on a Shopify theme that ships with an over-eager robots.txt blocking /collections/. The owner wonders why the category pages get 5 percent of the search traffic of the product pages. Robots.txt has been turning Googlebot away from the highest-value templates in the catalogue for three years.
A Joondalup professional services site that migrated from .com.au/oldsite to .com.au two years ago without any redirects. Half their old organic traffic still goes to a 404. The forensics turn up 200 backlinks pointing at pages that no longer exist. Three days of redirect work recover a measurable chunk of that authority.
A Bunbury healthcare practice running a React-based site with no server-side rendering. The doctor's bio pages, the only pages that should rank for branded queries, are invisible to Googlebot. A render-strategy change recovers them inside a fortnight.
None of these are exotic. They are the standard mistakes. The reason they persist is that nobody inside the business reads the robots.txt, nobody checks Search Console more than once a quarter, and the developer who set the site up has been gone for years. Technical SEO is the part of SEO with no internal owner in most Perth small businesses.
Tools and checklists worth using
You do not need an expensive stack. The minimum kit:
- Google Search Console. Free. The Coverage report, Crawl Stats, Sitemaps, and the Inspect URL tool cover most diagnostic needs. If you do not already have it set up, the chapter on crawling, indexing and ranking walks through it.
- Screaming Frog. Free up to 500 URLs, paid above. The single most useful crawler for technical SEO. Run a full crawl monthly. Sort by status code, indexability, canonical mismatch, hreflang errors, missing tags.
- PageSpeed Insights and Lighthouse. Free. Core Web Vitals data, render-blocking diagnostics, mobile usability checks. Run on your top 10 templates monthly.
- Your server logs. Free, if you can access them. The honest record of what Googlebot actually does on your site. Log file analysis for larger sites is where the real wins hide.
- The Schema.org Validator. Free. Paste your URL, check the structured data parses. Required reading before any site migration.
- Our free SEO audit tool. Pulls Lighthouse, Search Console-style indexability checks, schema validation and keyword data for any URL. Built to surface the technical issues this pillar covers. Run a free audit.
If you have a larger site, or a migration coming up, or e-commerce with thousands of SKUs, the human version of all of this lives in our website audit service and the ongoing SEO retainer. For sites where speed is the specific problem, speed optimisation sits alongside this pillar. For the operators in Perth specifically, Local SEO Perth is the local-flavoured starting point.
What to read next
Technical SEO is one of four interlocking pillars. The natural next steps depend on what you have already mastered:
- What is SEO? The foundational pillar. Start here if you want the big picture of how on-page, off-page and technical fit together. The chapter on how search engines work is the perfect prequel to this pillar.
- On-Page SEO pillar. If technical is the plumbing, on-page is what you put inside the house. The two pair tightly. URL structure and internal linking in particular straddle the technical-on-page line.
- Core Web Vitals pillar. Page experience is a technical sub-discipline that has grown to deserve its own pillar. If your audit flagged speed or layout-shift issues, that pillar covers them in full.
- Local SEO pillar. The location-based half of SEO. Pairs with this pillar when you have multi-location or multi-suburb pages to keep clean.
- AI Search pillar. Structured data, llms.txt and the new SERP layer. The technical groundwork in this pillar is what makes the AI-visibility work in that pillar possible.
- SEO Glossary. Plain-English definitions for every technical term in this guide.
- Free SEO audit. If you want the technical state of your own site checked before reading further, start here.