Technical SEO Audit Checklist

How to use this: work top to bottom the first time. The order is deliberate, because a rendering problem makes every speed metric below it unreliable, and an indexing problem makes everything else academic.

If a check fails, note it rather than fixing it immediately. Diagnose the whole site first, then fix in order of impact. Fixing as you go is how audits take three weeks and change nothing. When you have been through the list, the SEO audit report template turns those notes into a report someone can act on.

Crawling and indexing

robots.txt blocks nothing you need indexedOne stray Disallow can remove an entire section from search without any other symptom.
Your XML sitemap lists only canonical, indexable, 200-status URLsSitemaps full of redirects and 404s waste crawl budget and reduce trust in the file.
The sitemap is referenced in robots.txt and submitted in Search ConsoleDiscovery gets faster and the coverage report becomes genuinely useful.
No important page carries a noindex tag or X-Robots-Tag headerThis is the single most common reason a page silently disappears after a redesign.
The Pages report shows no unexpected exclusions"Crawled, currently not indexed" at scale is usually a quality or duplication signal, not a bug.
Crawl stats show no spike in server errors or response timeGoogle throttles crawl rate on sites that are slow or returning errors.
Faceted, filtered and session URLs are not crawlable at scaleParameter combinations can generate millions of near-identical URLs from a small catalogue.
Staging and development environments are not indexable and not linkedIndexed staging copies compete with production and leak unreleased work.

Site architecture and internal linking

Every important page is reachable within three clicks of the homepageDepth correlates with how often a page is crawled and how much internal authority reaches it.
No orphan pages: every indexable URL has at least one internal linkA page in the sitemap but linked from nowhere reads as unimportant.
Navigation uses crawlable HTML links, not JavaScript-only interactionsIf a menu only works on click in the browser, crawlers may never follow it.
Internal anchor text is descriptive rather than "click here" or "read more"Anchor text is one of the strongest signals for what a destination page is about.
Breadcrumbs are present on deep pages and marked upThey clarify hierarchy for users and can appear in the result snippet.
Pagination uses real crawlable links, not infinite scroll aloneContent only reachable by scrolling often never gets discovered.
Your highest-value commercial pages receive the most internal linksInternal linking is the one authority signal you fully control.

Rendering and JavaScript

Main content appears in the raw HTML, not only after JavaScript executesRendering is deferred and unreliable. Anything critical should not depend on it.
Critical links exist as real anchor tags in the initial responseLinks injected later may never be followed.
Rendered HTML in URL Inspection matches what a visitor seesA gap between the two is where most invisible ranking problems live.
No third-party script failure blocks rendering of primary contentOne slow tag manager can prevent the page from ever painting for a crawler.
Client-side routes each resolve to their own crawlable URLViews that never change the URL cannot be indexed independently.
Lazy-loaded content below the fold is not required for indexingContent that only loads on scroll or interaction is frequently missed.

Speed and Core Web Vitals

Largest Contentful Paint is under 2.5 seconds on mobile field dataLab tools flatter you. The 75th percentile of real visits is what counts.
Interaction to Next Paint is under 200 millisecondsINP replaced FID and is far less forgiving of heavy JavaScript.
Cumulative Layout Shift is under 0.1Late-loading images and injected banners are the usual culprits.
The hero image is preloaded and never lazy-loadedLazy-loading the largest above-fold element directly damages LCP.
Images are served in modern formats at the size actually displayedOversized images are the most common and most fixable weight problem.
Unused CSS and JavaScript is removed or deferredMost sites ship whole frameworks to render a brochure page.
Server response time stays under roughly 800 millisecondsEverything downstream inherits a slow first byte.

Duplication, canonicals and redirects

One page, one URL, one signal. Most of the damage here lands on eCommerce sites, where filters and variants multiply URLs faster than anyone audits them.

One canonical version of the site: HTTPS, single host, consistent trailing slashFour accessible variants of the same homepage split every signal four ways.
Every page has a self-referencing canonical unless deliberately pointing elsewhereMissing canonicals let parameters and variants create duplicates on their own.
Canonicals point to indexable 200 URLs, never to redirects or 404sA canonical to a redirect is treated as a hint at best and ignored at worst.
No redirect chain is longer than a single hopChains lose signal at every step and slow crawling.
301 is used for permanent moves rather than 302Temporary redirects tell Google to keep the old URL indexed.
Old URLs from previous builds redirect to their closest live equivalentRedirecting everything to the homepage throws away whatever those URLs had earned.
Paginated pages do not canonicalise back to page oneIt hides deeper items from discovery entirely.

Structured data

Machine-readable meaning, which now matters twice: once for rich results, and again for the AI answer engines now summarising your pages.

Organization or Person schema on the homepage with consistent namingThis is the anchor of your entity, and AI systems lean on it heavily.
Article, Product or Service schema matches what the page actually isMismatched types are ignored and occasionally penalised.
Breadcrumb schema is present and mirrors the visible breadcrumbsSchema that contradicts the page is worse than none.
No schema describes content that is not on the pageMarking up reviews or FAQs that users cannot see is a policy violation.
Rich Results Test passes with no errors on every templateTest one page per template rather than one page per site.

Mobile, security and page basics

The site is genuinely usable on a mid-range phone on a real connectionTest on hardware your customers own, not the newest device on your desk.
No content or links exist on desktop but are hidden on mobileGoogle indexes the mobile version. What is missing there is missing everywhere.
HTTPS is enforced everywhere with no mixed-content warningsA single insecure asset can strip the padlock from the whole page.
Tap targets and font sizes are usable without zoomingFrustration signals are cheap for Google to measure.
Each page has one H1 and a logical heading hierarchy underneath itHeadings are how both crawlers and assistants segment a page for extraction.

Serving more than one market

If regional versions exist, hreflang is reciprocal and self-referencingOne-directional hreflang is ignored, which is the most common implementation error.
No automatic geo-redirect traps crawlers in a single marketCrawling largely originates from one country. Force-redirect it and the rest never gets seen.
Currency, spelling and contact details do not contradict the market a page targetsMixed signals make it unclear which audience the page is for.
A single English site is not fragmented into thin near-duplicate regional pagesOne strong page usually outranks three thin ones split by country.

Verification and monitoring

Search Console and analytics are both installed and broadly agreeLarge unexplained gaps between them usually mean a measurement fault.
Conversion tracking attributes organic traffic correctlyIf you cannot attribute enquiries to search, you cannot justify the budget.
A crawl is scheduled regularly rather than run onceTechnical debt accumulates every time anyone ships.
Alerts exist for indexing drops and Core Web Vitals regressionsMost damage is discovered weeks late, and by then the cause is buried.
Every fix is logged with a date so impact can be attributed laterWithout a change log, you can never prove which fix moved anything.

Technical SEO Audit Checklist — Reportahasanadib.com

0/100

Your technical SEO health

Download your report as a PDF

Your score, your weakest area and the full numbered fix list, formatted as a report you can keep or hand to a developer. Add your details and the download unlocks on this page straight away.

Your details stay with me. No mailing list, no follow-up sequence, nothing passed on.

Rather have someone else run it?

A checklist tells you what to look at. It cannot tell you which nine of the fifty-four failures are actually holding your rankings down. That judgement is what a technical SEO audit buys you.

Book a free call →

Questions about this checklist

How often should a technical SEO audit be run?

A full pass once or twice a year for most sites, and after any migration, redesign or platform change. Sites that ship frequently need lighter monitoring monthly, because technical debt accumulates every time anyone deploys.

Do I need paid tools to work through this?

No. Search Console covers indexing, crawl stats and Core Web Vitals field data at no cost, and its URL Inspection tool answers most rendering questions. A crawler helps at scale, but every check here can be verified on a small site without one.

What is the difference between a technical SEO checklist and an audit?

A checklist tells you what to inspect. An audit adds the part a list cannot: working out which failures actually matter for your site, in what order, and what each one is worth fixing. A hundred flagged issues is not a plan.

Which check should I do first if I only do one?

Confirm nothing important carries a noindex tag. It takes two minutes, it is the most common cause of sudden ranking loss, and no other work on this list matters if the page cannot be indexed.