Technical SEO Audit Checklist (2026): 35 Checks, Free Template and Sample Report
By the PNext team · Updated 26 October 2026
A technical SEO audit answers one question: is anything stopping search engines from finding, understanding and trusting your pages? This checklist is the one we use, grouped by area, with a downloadable sheet and the layout we use for the final report.
When we ran it on our own website this month, it found 77 of 79 pages with two H1 tags, contact forms that had never sent an enquiry, and duplicate pages competing with each other. None of it showed up by just looking at the site, which is exactly why audits are worth doing.
- ✓Crawling and indexing checks
- ✓Speed and Core Web Vitals
- ✓Structured data and on-page HTML
- ✓International and AI crawler checks
- ✓What a good audit report looks like
What should a technical SEO audit include?
A technical SEO audit should check crawlability, indexing, site architecture, page speed and Core Web Vitals, mobile usability, structured data, canonicals and redirects, international tags such as hreflang, and access for search and AI crawlers. It should end with a prioritised list of fixes, not just a list of errors.
Crawling
Download: all 35 checks as a spreadsheet (CSV, opens in Excel or Google Sheets), with a status column for each check.
Crawl the whole site
Use a crawler such as Screaming Frog or Sitebulb and compare the result with your sitemap and analytics landing pages.
Check robots.txt
Make sure no important section is blocked by accident.
Find broken internal links
Every 404 linked from your own pages wastes crawling and frustrates visitors.
Find redirect chains and loops
Links should point straight at the final URL.
Check server errors
5xx responses, even occasional ones, can stop pages being crawled.
Look for crawl traps
Filters, calendars and parameters that create endless URLs.
Review log files if you can
They show what search engines actually crawl, not what you hope they crawl.
Indexing
Read the Pages report in Search Console
See which pages are indexed and the reasons others are not.
Check canonical tags
Each page should point to itself unless it is a deliberate duplicate.
Find accidental noindex tags
Especially on templates and after a redesign or staging launch.
Find duplicate and near-duplicate pages
Merge or redirect them so they stop competing.
Check the XML sitemap
Only indexable, canonical, 200-status URLs belong in it.
Check one URL version
https and one host (www or not), with the rest redirected.
Check thin pages
Tag, archive and search pages with little value should usually be noindexed.
Speed and Core Web Vitals
Check Core Web Vitals in Search Console
Field data for loading (LCP), responsiveness (INP) and stability (CLS).
Test key templates
Home, service, product and article templates with PageSpeed Insights.
Check image sizes and formats
Oversized images are the most common cause of slow pages.
Check render-blocking CSS and JavaScript
Defer what is not needed above the fold.
Check caching and server response
Slow server response holds back every other improvement.
Watch layout shifts
Scripts that move content after load can push CLS over the limit. We have seen this on our own pages.
On-page HTML and structured data
One H1 per page
Check the HTML, not the screen. Themes often add a hidden second H1.
Unique titles and descriptions
No duplicates across templates.
Image alt text
Descriptive alt text on meaningful images.
Structured data validity
Test with Google’s Rich Results Test and the Schema Markup Validator.
Organization or LocalBusiness schema
Consistent name, address and phone.
Breadcrumbs
Visible and in BreadcrumbList schema.
Internal linking
Important pages should be reachable within a few clicks and linked with descriptive anchors.
Mobile rendering
Content and links should match between mobile and desktop.
International SEO and AI crawlers
Check hreflang
Language and region codes are correct, and every alternate links back.
Check language URLs
Each language has its own URL; no automatic redirects by IP that hide versions from crawlers.
Check JavaScript rendering
Key content and links should appear in the rendered HTML Google sees.
Check AI search crawler access
Decide which AI crawlers to allow, such as OAI-SearchBot and PerplexityBot, and check robots.txt and firewalls match that decision.
Check llms.txt if you use one
Optional, but if present it should point to your best pages.
Check security headers and HTTPS
Valid certificate, no mixed content.
Check analytics and conversion tracking
Not strictly SEO, but an audit that cannot connect pages to leads misses the point.
What this checklist found on our own website
We ran every check above on performnext.com in October 2026 before publishing this guide. These are the real findings, what caused them, and how we fixed them. None of them were visible by just browsing the site.
| Check | What we found | Cause | Fix |
|---|---|---|---|
| One H1 per page (21) | 77 of 79 pages had two H1 tags | The theme printed a hidden page title above the H1 in our content | Turned off the theme title on pages that already have their own H1; rescan showed one H1 on all 79 |
| One URL version (13) | Inner pages loaded over plain http with a 200 status instead of redirecting | Only the homepage had an https redirect | One site-wide 301 rule to https and a single host, in one hop |
| Analytics and conversion tracking (35) | Contact forms on older pages showed a thank-you message but never sent anything | Forms were design placeholders with no handler | Routed every form to a working form and tested delivery |
| Layout shifts (20) | Cumulative Layout Shift of about 0.5 on some pages | A script adjusted spacing after the page had drawn | Moved the spacing into CSS; CLS back to 0 |
| Image sizes and loading (17) | Mobile LCP around 2.9 seconds | The header logo was lazy-loaded even though it is the first thing on screen | Excluded the logo from lazy-loading and deferred non-critical scripts; LCP about 2.7 seconds |
| Duplicate pages (11) | Copies of service pages with “-usa” in the URL competing with the real USA pages | Leftovers from an earlier build | 301 redirects to the correct page and removal from the sitemap |
| Server errors (5) | Google’s PageSpeed test got a 503 error on one page | Shared hosting limits under load | Static page caching so most visits never reach PHP |
Two lessons from doing this on our own site. First, the most damaging issue, forms that did not send, was not an SEO problem at all, which is why check 35 is on the list. Second, several problems came from the theme and plugins rather than anything we wrote, so an audit has to read the HTML Google receives, not just the page you see.
What a good technical SEO audit report looks like
A useful report is short and ranked. This is the structure we use:
| Section | What it contains |
|---|---|
| Summary | The five issues that matter most, in plain English, with expected impact |
| Prioritised fix list | Every issue with impact, effort, affected URLs and who should fix it |
| Evidence | Screenshots, crawl exports and Search Console data behind each finding |
| Quick wins | Fixes that take under a day and can go live this week |
| Monitoring plan | What to check after the fixes and when |
If you would like this done for you, our technical SEO services deliver exactly this report, and for large sites see what enterprise SEO involves. The broader SEO checklist covers content and links as well.
Want a technical SEO audit?
We will run all 35 checks on your site and send a prioritised report your developers can act on.
The checks by priority
Not all 35 checks matter equally. If you only have an afternoon, work through them in this order:
| Priority | Checks | Why first |
|---|---|---|
| Critical | Important pages blocked by robots.txt or noindex; wrong canonicals; 5xx errors; sitemap missing key pages | If Google cannot crawl or index a page, nothing else matters |
| High | Broken internal links and redirect chains; duplicate titles; slow templates (Core Web Vitals); mobile issues | Wastes crawl and loses users and rankings |
| Medium | Structured data errors; thin or duplicate pages; orphan pages; hreflang | Helps search engines understand and rank pages |
| AI crawlers | robots.txt rules for GPTBot, ClaudeBot, PerplexityBot and Google-Extended; an llms.txt file; server-rendered content | Decides whether AI assistants can read and cite your pages |
Running a migration? Use our SEO migration checklist alongside this one. Planning the fixes? Our SEO roadmap template helps you order them.
Further reading from our team: SEO proposal template with a filled example.
Frequently Asked Questions
Try a different keyword, or ask our team directly.
A small site can be audited in a few days. Large sites with many templates or markets take two to four weeks, including the report.
A full audit once or twice a year, plus a quick check after every redesign, migration or major release.
A crawler such as Screaming Frog or Sitebulb, Google Search Console, PageSpeed Insights, the Rich Results Test and your analytics. Log file analysis helps on larger sites.
A technical audit covers how search engines crawl, render and index the site. A full SEO audit also covers content, keywords and links.
It depends on your goals. Blocking AI search crawlers can keep you out of AI answers. Many businesses allow search-focused crawlers and make a separate decision about training crawlers.
Anything that stops important pages being crawled or indexed, then issues on templates that affect many pages, then individual pages.
More Guides for US Businesses
Ready to talk about your results?
Tell us about your site and we will send a prioritised audit your team can act on.
We work with businesses across the USA and the UAE. No pitch, no pressure, just an honest view of what we would do.
Or call and WhatsApp us on +971 50 298 5401.
- 01Crawl and indexing
- 02Speed and Core Web Vitals
- 03Schema and HTML
- 04International and AI crawlers
Get a technical SEO audit
Tell us about your website and goals.