What a technical SEO audit covers: the eight areas, the errors that cost most, and the 2026 checklist
A technical SEO audit finds what stops Google crawling, indexing and showing a site. The eight areas, the costly errors, and a checklist with pass conditions.
A technical SEO audit checks whether Google can crawl your site, index the right version of each page, render it on a phone and read its structured data. It does not judge whether the pages are any good. That is a separate job, and it only pays off once the technical side is sound.
Speed is part of it. Google publishes three Core Web Vitals targets, measured on real visits. They are Largest Contentful Paint within 2.5 seconds, Interaction to Next Paint of 200 milliseconds or less, and Cumulative Layout Shift of 0.1 or less (web.dev). Google says these are used by its ranking systems, and also that good scores do not guarantee top rankings (Google Search Central). Below is what an audit covers, the checklist with a pass condition for each check, and the errors that cost the most.
The short version
- A technical SEO audit checks crawling, indexing, rendering, speed, mobile, security, structured data and international setup. It does not judge content quality.
- Robots.txt controls crawling, not indexing. A page blocked there can still appear in Google without a description, and a page you want out of results needs a noindex rule that Google can crawl (Google Search Central).
- A canonical tag is a hint, not a rule. Google may choose a different canonical, so redirects, canonicals, sitemaps and internal links should all point at the same URL (Google Search Central).
- Google indexes and ranks the mobile version of a site, so the phone view is the one to audit (Google Search Central).
- Poor page experience can hold a page back, but Google says there is no single page experience signal and relevance still comes first.
- Crawl budget is mainly a concern for very large or fast-changing sites. Google’s guide is written for sites with 10,000 or more pages that change daily, or a million or more that change weekly.
What is a technical SEO audit, and what is it not?
A technical audit examines how search engines crawl, index and interpret the site. It does not tell you whether a page deserves to rank. A site can pass every technical check and still get no search traffic because it has no page for what customers search. A site with excellent pages can stay invisible because one line in one file told Google to stay out. The audit removes the second problem so the first one can be worked on.
No single tool covers everything. A real audit combines a crawler such as Screaming Frog or Sitebulb with Search Console and a person reading the output. The free version of Screaming Frog crawls up to 500 URLs (Screaming Frog), which is enough for many small business sites. A crawler reports hundreds of findings. A person decides which five matter.
The technical SEO audit checklist
Priority here is our recommendation for a typical small or mid-sized site, not a Google ranking. Work down the High rows first.
| Area | Check | How to check | Pass condition | Priority |
|---|---|---|---|---|
| Crawling | robots.txt does not block pages you want found | Open yourdomain.com/robots.txt and read every Disallow line | No Disallow covering a page, folder, CSS or JavaScript file you need crawled | High |
| Indexing | Important pages are indexed | Search Console, Pages report, then URL Inspection on key pages | Every page you want to rank shows as indexed | High |
| Indexing | No stray noindex | View source and search for noindex, or read URL Inspection | No noindex on any page you want in results | High |
| Indexing | noindex pages are crawlable | Compare noindexed URLs against robots.txt | No page relies on noindex while also being blocked in robots.txt | Medium |
| Status codes | Live pages return 200, removed pages 404 or 410 | Crawler status code report | No internal links to 4xx or 5xx pages, no soft 404s in Search Console | High |
| Redirects | No chains or loops | Crawler redirect chain report | Each old URL reaches its final page in one hop | Medium |
| Canonicals | Signals agree | Crawler canonical report, URL Inspection “Google-selected canonical” | Canonical, redirects, sitemap and internal links name the same URL | High |
| Sitemap | Sitemap lists only pages you want indexed | Search Console, Sitemaps report, and a crawl of the sitemap URLs | Indexable, canonical, 200-status URLs only, under 50,000 URLs per file | Medium |
| Structure | Important pages are easy to reach | Crawler crawl depth report | Mavix recommendation: key pages within three clicks of the homepage | Medium |
| Structure | No orphan pages | Compare crawler URLs with sitemap and analytics URLs | Every page you want to rank has at least one internal link | Medium |
| Rendering | Main content is in the HTML | View source and search for a sentence from the page | The text is there without JavaScript running | Medium |
| Speed | Core Web Vitals pass on real visits | Search Console Core Web Vitals report, PageSpeed Insights field data | LCP 2.5s or less, INP 200ms or less, CLS 0.1 or less at the 75th percentile | High |
| Mobile | Phone view carries the same content | Load key pages on a real phone | Same main content, links and structured data as desktop, no intrusive interstitials | High |
| Security | HTTPS everywhere | Load http:// versions, check the browser console | Every HTTP URL redirects to HTTPS, no mixed content warnings | High |
| Structured data | Markup is valid and matches the page | Rich Results Test on each template | No errors, every marked-up value visible on the page | Medium |
| International | hreflang is correct, if used | Crawler hreflang report | Each language version lists itself and all others, with return links | Low unless multilingual |
| Local | Business details match | Compare site footer, schema and Google Business Profile | Same name, address and phone number everywhere | Medium for local firms |
A longer version of this list, as a spreadsheet you can fill in, is the SEO audit template.
What are the eight areas an audit covers?
The table above is the working list. This is what each area means and why it is there.
Crawlability and indexation. Search Console shows no critical crawl errors, robots.txt blocks the right things and nothing else, the sitemap is submitted and clean, canonical tags are set, and URL variations do not create duplicates.
Site structure and URLs. Clean readable URLs, breadcrumbs, internal links that follow the logic of the site, and no orphan pages. Our recommendation is that key pages sit within three clicks of the homepage. That is a Mavix rule of thumb for small and mid-sized sites, not a Google requirement.
Speed and Core Web Vitals. The three thresholds above, read from field data rather than a lab score. Compressed images in modern formats, deferred scripts, and a content delivery network where it helps. The Core Web Vitals thresholds post has every boundary.
Mobile. Google uses the mobile version of a site’s content, crawled with a smartphone agent, for indexing and ranking (Google Search Central). Tap targets sized for a thumb, no interstitials covering the content, the viewport set correctly.
Security. Everything on HTTPS, no mixed content, a certificate that renews itself, and a permanent redirect from every HTTP address to its HTTPS twin. Google prefers HTTPS pages over equivalent HTTP pages as canonical (Google Search Central).
Structured data. Schema for the organisation, the local business, articles and FAQs where they exist, with no errors in the Rich Results Test, and values that match what a visitor can see.
Local signals. A complete Google Business Profile linked to the site, the same name, address and phone number everywhere, and local business schema.
International. hreflang only if the site serves more than one language or region. Each language version must list itself and all the others, and if two pages do not point at each other the tags may be ignored (Google Search Central).
Photo: Mikhail Nilov on Pexels.
Which technical SEO errors cost the most?
The errors below are the ones that can keep whole sections of a site out of Google, which is why they come first. Each has a documented cause and a plain fix.
A robots.txt line that blocks the site. Disallow: / under User-agent: * asks every crawler to stay out of the whole site (Google Search Central). The file controls crawling rather than indexing. A page blocked there can still appear in results with no description, and a page you want out of results needs a noindex rule instead. Google also warns against blocking CSS and JavaScript when the page is harder to understand without them (Google Search Central).
The WordPress visibility box. Under Settings, then Reading, WordPress has a box labelled “Discourage search engines from indexing this site”. Since version 5.3, ticking it adds a noindex, nofollow robots meta tag to every page (WordPress.org). It gets ticked during development and sometimes left on after launch. WordPress runs 40.2 percent of all websites, according to W3Techs on 30 September 2026 (W3Techs). So it is the first box to check on a new WordPress site that gets no search traffic.
Canonicals that disagree with everything else. HTTP and HTTPS, www and non-www, trailing slash and none, and filter parameters can each produce a copy of every page. Google treats a canonical preference as “a hint, not a rule”, and may choose a different page (Google Search Central). Redirects and rel=“canonical” are strong signals, sitemap inclusion is a weak one, and they work best combined (Google Search Central). Make the canonical, the redirects, the sitemap and the internal links name the same URL.
Redirect chains. Page A sends to B, which sends to C. Google’s crawlers follow up to 10 hops by default (Google Search Central). Google advises keeping chains short, ideally no more than three, because each hop adds latency (Google Search Central). Long chains also have “a negative effect on crawling” on large sites (Google Search Central). Find them with a crawler and point every old address straight at its final page.
A sitemap full of pages you do not want indexed. Google says to include the URLs you want to see in search results (Google Search Central). So keep it to canonical, indexable pages that return a 200 status: no redirects, no noindexed pages, no cart or thank-you pages. Wrong entries send mixed signals and waste crawling on pages that are not meant to rank. Update the lastmod date only after a significant change to the page, because that is what Google says the value should reflect.
Mixed content after an HTTPS move. An HTTPS page that loads images or scripts over HTTP has mixed content. Browsers block the risky kinds, such as scripts, and warn about or upgrade the rest (web.dev). Chrome’s DevTools Issues tab lists each one. Update every address to HTTPS.
Structured data that does not match the page. Google’s rule is not to mark up content that is not visible to readers (Google Search Central). Review markup claiming five hundred reviews on a page that shows ten breaks it. A structured data manual action removes the page’s eligibility for rich results, and Google says it does not affect how the page ranks in web search.
Plugin and script weight. Some WordPress plugins load their scripts and styles on every page, including pages that do not use them. That extra code can push Largest Contentful Paint or Interaction to Next Paint over the thresholds. Test each plugin’s effect on a staging copy, add caching and image compression, and remove what nobody can explain.
Does crawl budget matter for a small site?
For most small sites, no. Google’s crawl budget guide is written for large sites with a million or more unique pages that change about weekly. It also covers sites with 10,000 or more pages that change daily (Google Search Central). Google says that if your site does not have a large number of rapidly changing pages, you do not need the guide.
On a large catalogue it does matter, and the fixes are the ones above. Block parameter duplicates, remove soft 404s, which Google says keep getting crawled, fix broken internal links, and keep the sitemap clean.
Photo: Brett Sayles on Pexels.
How much does a technical SEO audit cost?
It depends on the size of the site and the depth of the work. Five brochure pages are a fraction of the job of a two-thousand-product catalogue with log file analysis and JavaScript rendering checks. What you are paying for is not the list of findings, which any crawler produces. It is the prioritised plan: what to fix first, what it takes, and what it should change. A spreadsheet of nine hundred warnings is not an audit. The technical search engagement here sets out the scope and what drives the figure.
What will fixing technical problems change?
It depends on what was broken. If a page was blocked, noindexed or unreachable, fixing that can bring a visible change, because the page could not appear before. If the site was already crawlable and just slow, the effect is smaller. Google says its systems always look for the most relevant content, even when page experience is below par (Google Search Central). Poor page experience can hold a page back against close competitors. It does not set a limit that relevance and quality cannot pass.
La Vela Coffee, an account managed here, went from 1.16K to 2.85K organic clicks in six months against the six before, according to its Search Console comparison. The technical work was the first fortnight. The rest was pages written for what its customers searched. The report does not name calendar dates, so read it as one account’s result, not a rate to expect.
Where the line is
An audit fixes nothing on its own. Its value depends on whether the findings get implemented, and often they do not, because the report went to an owner and the fixes needed a developer. Agree who will do the work before you commission the audit.
Our recommendation is to run one once a year, and again after a redesign, a migration, a platform change or an unexplained drop. The website launch checklist covers the redesign case.
The bottom line
A technical audit answers one question: can Google reach, read and trust every page that matters, on a phone, today? When the answer is yes, the site has stopped getting in its own way, and the work on pages people want to read can start paying off. This guide explains what an audit checks and why. To run one yourself, use the 42-point SEO audit checklist, or download the SEO audit template with a pass condition for every row.
Questions people ask about this
How do I check Core Web Vitals on real visits?
Open Search Console and look at the Core Web Vitals report, which uses field data from real Chrome users. It groups URLs into good, needs improvement and poor for mobile and desktop. PageSpeed Insights shows the same field data at the top of its report for a single page, above the lab score. If the field section says there is not enough data, your site has too few visits to measure, and the lab score is the only guide you have.
Can I run the audit myself?
Yes, for most of it. Search Console covers crawl errors, indexing, Core Web Vitals and mobile problems for free. A crawler such as Screaming Frog is free for up to 500 URLs and finds redirect chains, broken links, missing canonicals and orphan pages. What you cannot easily do yourself is judge which findings matter and what to fix first, which is the actual product. Start with the free tools and see how far you get.
What is the difference between noindex and robots.txt?
Robots.txt tells Google not to crawl a page. Noindex tells Google not to show it in results. They are not interchangeable. A page blocked in robots.txt can still appear in results without a description, because Google saw links to it. A page with noindex must be crawlable so Google can read the tag. If you want a page out of results, use noindex and leave it open in robots.txt.
How do I find orphan pages?
Compare two lists: pages the crawler found by following links, and pages that exist according to the sitemap, Search Console and analytics. Anything in the second list and not the first is an orphan. Screaming Frog and Sitebulb both do this comparison if you connect Search Console and analytics before crawling. Then decide for each one: link it from a relevant page, redirect it, or noindex it if it should never have existed.
Do I need a CDN for a small site?
Not always. A content delivery network helps most when visitors are far from the server or the pages carry heavy images. A local business with a small site on a decent host may pass the speed thresholds without one. Check the field data first. If Largest Contentful Paint fails, compress the images and defer scripts before paying for anything. If it still fails, a CDN is the next step, and many hosts include one.
What should I do after a site migration?
Before switching, map every old address to its new one and set permanent redirects for each, straight to the final page. After switching, resubmit the sitemap, check Search Console for a spike in errors, crawl the old addresses to confirm they redirect once, and search for mixed content. Watch impressions daily for two weeks. A dip of a few weeks is normal. A drop that keeps going means addresses were missed.
How many plugins is too many on WordPress?
The count matters less than what each one loads on every page. Deactivate one plugin at a time on a staging copy and re-test a page. The ones that shift Largest Contentful Paint are the problem. Keep one for caching, one for images and one for forms, and question everything else. A plugin nobody can explain the purpose of should go. Then re-check the field data a month later, because that is what real visitors experienced.
Will fixing technical issues raise my rankings?
It depends on what was broken. If a page was blocked, noindexed or unreachable, fixing that can produce a fast and visible change, because the page was never in the race. If the site was already crawlable and just slow, the change is smaller and slower. Pair the fixes with pages written for what customers search, or the audit produces a clean site nobody visits.
How do I test my structured data?
Paste the page address into Google’s Rich Results Test and read the result. It reports errors, warnings and which rich result types the page is eligible for. Then check by eye that every value in the markup appears on the visible page: the rating, the count, the price, the address. Markup that says something the visitor cannot see is the violation described above, and it is common on review plugins.
What does a redirect chain look like in practice?
An old blog address that goes to an HTTPS version, then to a version with a trailing slash, then to the new page. Three hops where one would do. Crawlers list these under redirect chains, and you can spot them in the network tab of your browser as several requests before the page appears. Fix each by pointing the first address straight at the final page, and update internal links so they no longer use the old address.
Should I audit before or after a redesign?
Both, and a redesign is named above as one of the triggers. Before, so you know which pages carry traffic and links and must keep their addresses or be redirected. After, because redesigns break canonicals, drop redirects, reintroduce the WordPress noindex box and add scripts that fail the speed thresholds. Run the before audit early enough to shape the build, and the after audit in the first week live.
Free review
Want this looked at on your own site?
Send the address. A person reads it and writes back with the two or three things that would move your enquiry number first. Free, and yours to keep whether you hire us or not.
