Technical SEO is the work of making sure a search engine can reach, render, understand and store your pages — and almost every real failure happens in the first two of those, not in the sophisticated parts people spend their time on.
The useful mental model is a stack of four layers, each of which must succeed before the next one is even attempted:
| Layer | The question | What failure looks like |
|---|---|---|
| Crawl | Can Googlebot fetch this URL? | Nothing happens at all. No impressions, no data |
| Render | Can it see the content once fetched? | Page indexed with no content, or missing sections |
| Index | Did it decide to store the page? | "Crawled — currently not indexed" |
| Rank | Where does it place against competitors? | Indexed, but on page four |
Most technical SEO advice addresses layer four. Most technical SEO problems live in layers one and two. That mismatch is why sites get audited repeatedly and stay broken.
The failures that actually cost traffic
In rough order of how often they turn out to be the answer.
A robots.txt rule blocking something important. Usually inherited from a staging environment and never removed. This is the highest-consequence, lowest-difficulty failure in the entire discipline, and it is covered in robots.txt mistakes that de-index whole sites.
A noindex left in place after launch. Same origin, same cause. WordPress has a "Discourage search engines" checkbox that does exactly this, and it survives launches routinely.
Content that only exists after JavaScript runs. Google renders JavaScript, with a delay and not always completely. If your main content requires a client-side fetch to appear, you are gambling on rendering succeeding every time.
Canonical tags pointing at the wrong URL. A canonical pointing to the homepage from every page is a surprisingly common template bug, and it tells Google that none of those pages deserve to exist separately.
Redirect chains and loops. Every hop loses a little, and long chains get abandoned.
Server errors under crawl load. A host that returns 5xx when Googlebot arrives in volume teaches Google to crawl less.
Duplicate URLs from parameters and pagination. Filters, session IDs, tracking parameters and print views quietly multiply one page into forty.
None of these are exotic. They are the same handful of failures on almost every site, and a diagnostic that checks them in order finds the problem faster than any audit tool's 200-point report.
Core Web Vitals, honestly
Three metrics, with Google's stated targets:
| Metric | Measures | Target |
|---|---|---|
| LCP — Largest Contentful Paint | Loading | Under 2.5 seconds |
| INP — Interaction to Next Paint | Responsiveness | Under 200 milliseconds |
| CLS — Cumulative Layout Shift | Visual stability | Under 0.1 |
INP replaced First Input Delay, so any guidance still telling you to optimise FID is out of date.
Google's own framing is worth reading carefully: good Core Web Vitals "align with what our core ranking systems seek to reward." That is deliberately weaker than "these are ranking factors," and the practical translation is:
Core Web Vitals will not rescue a page that does not deserve to rank, and they will lose you ground against a competitor who is otherwise equal. They are a tiebreaker, and they matter most where competition is close.
The trap to avoid: chasing a perfect Lighthouse score. Lighthouse is a lab simulation. Search Console reports field data from actual visitors, and field data is what counts. A site can score 100 in Lighthouse and fail Core Web Vitals in the field, usually because real users are on worse connections and slower devices than the simulation assumes.
What matters less than people think
Keyword density, meta keywords, and exact-match anything. These have not been meaningful for many years and still appear in audit templates.
Perfect scores on third-party audit tools. They flag issues by rule, not by consequence. A tool reporting 340 "errors" on a healthy site is measuring its own ruleset.
XML sitemap perfection. Sitemaps help discovery on large or poorly linked sites. They do not cause indexing, and a page that a sitemap lists but nothing links to is still an orphan.
Schema volume. More markup is not more signal. And two schema types most sites were told to add now produce nothing: HowTo rich results were withdrawn in 2023 and FAQ rich results stopped appearing on 7 May 2026. Neither type appears in Google's structured data gallery any more. What is still supported, and what a blog actually needs, is in schema markup for blogs.
Crawl budget, for most sites. It is a genuine constraint above roughly a hundred thousand URLs. Below that, a site worrying about crawl budget almost always has a different problem wearing its clothes.
The audit order
Seven steps. Stop at the first thing that is broken, fix it, and re-check before continuing — because a failure at one layer produces misleading symptoms at every layer above it.
1. Fetch your robots.txt and read it. Not with a tool. Open yourdomain.com/robots.txt and read every line. This takes thirty seconds and finds the most expensive problem more often than any other single check.
2. Run URL Inspection on your five most important pages. Search Console will tell you whether each is indexed, what the canonical is, and whether the live version can be fetched. This is the single most useful tool Google gives you.
3. Read the Page Indexing report by reason. Not the total — the reasons. Which category is growing tells you what kind of failure you have. The full decode is in why isn't my page indexed.
4. View one page with JavaScript disabled. If the main content disappears, you have a rendering dependency and should know it deliberately rather than discover it later.
5. Check canonicals on a template basis, not page by page. Canonical bugs are almost always template-wide, which means one wrong tag affects thousands of URLs.
6. Look at Core Web Vitals field data in Search Console, not a Lighthouse score.
7. Check the AI performance report. New as of 31 August 2026 — impressions for AI Overviews, AI Mode and generative AI in Discover, separated from ordinary Search. Impressions only, no clicks. It is the newest thing in Search Console and the least likely to be in an audit template.
Which reports repay regular attention, and which to ignore, is in Search Console: the 5 reports that matter.
When "technical SEO" is not the problem
Worth saying plainly, because it saves money.
If your pages are indexed and simply not ranking, you do not have a technical problem. You have a competitive one, and no amount of schema, sitemap tuning or Core Web Vitals work will change it. Indexed-but-not-ranking is a content and authority question.
The tell is simple: technical problems produce zero, not low. A page with no impressions has a technical problem. A page with impressions and few clicks has a relevance or presentation problem. A page with clicks and no conversions has a content problem.
Reading those three symptoms correctly is most of the diagnostic skill in this discipline.
Frequently asked questions
What is technical SEO?
The work of making sure search engines can reach, render, understand and store your pages. It covers four layers — crawl, render, index, rank — and almost all real failures happen in the first two rather than in the more sophisticated work people usually spend time on.
What technical issues actually hurt rankings most?
A robots.txt rule blocking something important, a noindex left in place after launch, content that only appears after JavaScript runs, canonical tags pointing at the wrong URL, redirect chains, server errors under crawl load, and duplicate URLs from parameters. The same handful on almost every site.
Do Core Web Vitals affect rankings?
Google says good Core Web Vitals align with what its ranking systems reward, which is deliberately weaker than calling them ranking factors. In practice they are a tiebreaker: they will not rescue a page that does not deserve to rank, and they can cost ground against an otherwise equal competitor.
What are the current Core Web Vitals thresholds?
Largest Contentful Paint under 2.5 seconds, Interaction to Next Paint under 200 milliseconds, and Cumulative Layout Shift under 0.1. INP replaced First Input Delay, so guidance still referring to FID is out of date.
Why is my page indexed but not ranking?
That is not a technical problem. Technical failures produce zero impressions, not low rankings. An indexed page with impressions and poor position is competing on relevance, content quality and authority, none of which schema or sitemap work will change.
Does schema markup improve rankings?
Not directly. It helps machines understand a page and can produce rich results for supported types. Two types widely recommended for years now produce nothing — HowTo rich results were withdrawn in 2023 and FAQ rich results stopped appearing on 7 May 2026, and neither appears in Google's structured data gallery.
Do I need to worry about crawl budget?
Almost certainly not below roughly a hundred thousand URLs. It is a real constraint on very large sites. On smaller ones, a site that believes it has a crawl budget problem usually has a different problem — orphaned pages, duplicate URLs, or server errors — wearing its clothes.
Should I aim for a perfect Lighthouse score?
No. Lighthouse is a lab simulation on assumed hardware and connection. Search Console reports field data from actual visitors, and field data is what Google uses. Sites routinely score 100 in Lighthouse while failing Core Web Vitals in the field.
What to do next
Open yourdomain.com/robots.txt and read it, line by line. Thirty seconds, no tools, and it finds the most expensive class of problem more often than anything else you could do first.
Then run URL Inspection on your five most important pages and confirm each one is actually indexed with the canonical you expect.
Related guides
- Why isn't my page indexed? A diagnostic — decoding the Page Indexing report
- robots.txt mistakes that de-index whole sites — the highest-consequence failure
- Schema markup for blogs: what to add and how — what still produces rich results
- Search Console: the 5 reports that matter — where the answers are
- SEO when AI answers the question first — what changed above the technical layer
Chapters in this Guide
Free: The 60-Minute Email Authentication Fix
A no-fluff checklist to set up SPF, DKIM & DMARC correctly and pass Gmail & Yahoo's sender requirements.
Muhammad Basim has worked in digital marketing since 2013, focused on email deliverability and AI-assisted content production. He is the author of The Email Deliverability Playbook and The Email Copywriting Playbook.
Related Articles
Email Deliverability: Why Authenticated Emails Still Land in Spam
Email deliverability is whether your message reaches the inbox rather than the spam folder, and it is decided by three factors: technical infrastructure, list quality, and sending behaviour. Authentication is one item inside the first factor. Getting it right is necessary, and it is nowhere near sufficient. That gap explains the most common complaint in […]
Linkable Assets: What Actually Earns Links
A link is a citation, and a citation requires that someone writing about your subject needed you to make their point. That is the whole mechanism, and it is the reason most "linkable content" earns nothing. The test, before you build anything: Could a writer covering this topic finish their sentence without referencing you? If […]
Link Outreach Emails That Get Replies
Outreach fails for exactly two reasons, and only one of them gets written about. The first is that the email was not worth replying to. That is the reason every guide addresses, and the advice — personalise, be brief, offer value — is correct and insufficient. The second is that the email never arrived. Nobody […]