Technical SEO is the foundation every other SEO investment depends on. You can produce excellent content and earn strong backlinks — and rank nowhere if Googlebot can’t crawl your pages, your Core Web Vitals fail, or your JavaScript-rendered content never gets indexed.
This 40-point checklist is organized by category. For each item: what it is, the most common mistake, and what correct looks like.
Crawlability
1. robots.txt is correctly configured.
Googlebot reads robots.txt before crawling. The most common mistake: accidentally blocking / in a staging robots.txt that gets copied to production. Check: https://yourdomain.com/robots.txt and verify no critical paths are disallowed.
2. Crawl budget is not wasted on low-value URLs.
Sites with URL parameters, faceted navigation, or session tokens can generate thousands of duplicate or near-duplicate URLs. Common mistake: allowing parameterized URLs into the index (e.g., /products?color=blue&sort=price). Fix: disallow parameter variants in robots.txt or use canonical tags.
3. XML sitemap is present, accurate, and submitted. Sitemaps tell Google which URLs you want crawled. Common mistake: the sitemap includes noindex pages, redirected URLs, or pages returning 4xx/5xx errors. Run a sitemap audit to ensure every URL in the sitemap is crawlable and indexable.
4. Sitemap is linked from robots.txt.
Sitemap: https://yourdomain.com/sitemap.xml should appear in robots.txt. Common mistake: relying only on Search Console submission without the robots.txt reference.
5. No unnecessary redirect chains. A redirect chain (A → B → C) wastes crawl budget and dilutes PageRank. Common mistake: multiple migrations stacked on top of each other. Every redirect should go directly to the final destination URL. Audit with a crawl tool and collapse all chains.
6. Response time for key pages is under 200ms. Slow server response doesn’t just hurt user experience — it reduces how many pages Googlebot can crawl per session. Common mistake: database queries on every request without caching.
7. No soft 404s. A soft 404 returns a 200 status code for a page that displays a “not found” message. Common mistake: CMS setups that return 200 even for deleted product or post pages. Google eventually identifies these but they waste crawl budget and may prevent real pages from being discovered.
Indexation
8. Canonical tags are implemented correctly. The canonical tag tells Google which URL is the authoritative version of a page. Common mistake: self-referencing canonicals with the wrong domain (HTTP vs HTTPS, www vs non-www), or paginated pages all pointing to page 1.
9. noindex is not applied to pages you want ranked.
Check every noindex tag in your templates. Common mistake: a WordPress or CMS plugin that adds noindex to categories, tags, or archive pages — all of which should often be indexed.
10. hreflang is correctly implemented for multilingual sites. hreflang tells Google which language/region each URL serves. Common mistake: hreflang annotations that don’t reference each other reciprocally (every page in the hreflang set must reference all other pages in the set).
11. Pagination is handled correctly.
Since Google deprecated rel="next/prev", the recommended approach is strong canonical tags on paginated pages and ensuring the first page is the strongest entry point. Common mistake: paginated pages with no canonical and no internal links, floating disconnected from the site.
12. Duplicate content is resolved. Near-duplicate pages dilute authority. Common mistake: www and non-www, HTTP and HTTPS versions of the site all serving content without a redirect or canonical. Ensure a single canonical URL serves each piece of content.
Core Web Vitals
13. LCP (Largest Contentful Paint) is under 2.5 seconds.
LCP measures when the largest visible element renders. Common mistake: hero images served without loading="eager" or preload hints, causing the browser to discover them late. Use <link rel="preload" as="image"> for above-the-fold images.
14. INP (Interaction to Next Paint) is under 200ms. INP replaced FID in March 2024. It measures the delay between user interaction and visible response. Common mistake: heavy JavaScript blocking the main thread during user interactions. Defer non-critical scripts and move expensive tasks off the main thread.
15. CLS (Cumulative Layout Shift) is under 0.1. CLS measures unexpected visual instability. Common mistake: images and ads without explicit width/height dimensions, causing layout reflow when they load. Always declare image dimensions in HTML.
16. Server response time (TTFB) is under 800ms. Time to First Byte affects all other metrics. Common mistake: no edge caching for static or semi-static pages. Use a CDN with appropriate cache headers.
17. Core Web Vitals are measured in field data, not just lab data. Search Console’s Core Web Vitals report shows real user data. Common mistake: relying only on PageSpeed Insights lab scores while field data (which Google actually uses for ranking) shows failing URLs.
Mobile and HTTPS
18. The site is mobile-responsive and passes Mobile Usability test. Google uses mobile-first indexing. Common mistake: a desktop layout that renders poorly on small viewports — text too small, tap targets too close together. Check Search Console’s Mobile Usability report.
19. All pages are served over HTTPS. HTTPS is a confirmed ranking signal. Common mistake: mixed content — pages served over HTTPS but loading HTTP resources (images, scripts). Browser DevTools Network tab will flag these.
20. HTTPS redirects are permanent (301), not temporary (302). Temporary redirects don’t pass PageRank reliably. Common mistake: server misconfiguration using 302 for the HTTP-to-HTTPS redirect.
Structured Data
21. Article schema is implemented on blog and news content.
Article or NewsArticle schema enables rich results. Common mistake: missing author, datePublished, or headline properties — all of which are required or strongly recommended.
22. FAQ schema is implemented on pages with Q&A sections. FAQ schema can earn expanded rich results in SERPs. Common mistake: implementing FAQ schema on pages where the Q&A content is not visible to users — Google may penalize hidden structured data.
23. Product schema includes price, availability, and reviews. For e-commerce, Product schema enables rich results including price, star ratings, and availability in the SERP. Common mistake: price not updated dynamically when it changes — stale structured data.
24. LocalBusiness schema on all location pages.
LocalBusiness schema with name, address, telephone, openingHours, and geo coordinates gives Google the data it needs to serve local results. Common mistake: missing geo coordinates or openingHours — both heavily used in local ranking.
25. Schema markup validates without errors. Test all schema implementations via Google’s Rich Results Test and Schema Markup Validator. Common mistake: implementing schema in the head or body without verifying it renders in the HTML source after JavaScript execution.
JavaScript SEO
26. Critical content is not hidden behind JavaScript. Googlebot can execute JavaScript, but with delays and inconsistency for complex SPAs. Common mistake: putting primary body content, headings, and internal links inside client-side JavaScript components that render only after page load — Google may never see them.
27. Internal links are crawlable anchor tags, not JavaScript event handlers.
Google follows <a href="..."> tags. Common mistake: navigation built with JavaScript event listeners (onClick) rather than real href attributes — Googlebot can’t discover the linked pages.
28. Server-side rendering or static generation is used for content pages. If you’re using a JavaScript framework (React, Vue, Next.js), ensure content pages are SSR or SSG, not client-side only. Common mistake: a Next.js site where blog posts are rendered client-side — Google may index blank shells.
29. Lazy-loaded content uses Intersection Observer correctly. Google can trigger Intersection Observer during rendering, but not always reliably. Common mistake: lazy-loading all body content — including primary text — beneath the fold, with no server-rendered fallback.
Internal Linking
30. No orphan pages exist in the index. Orphan pages have no internal links pointing to them — Google discovers them only from the sitemap, and they rarely rank. Common mistake: publishing new content without adding links from relevant existing pages.
31. Link equity flows to high-priority pages. Your homepage has the most PageRank. Internal links distribute that equity downward. Common mistake: deep pages receiving no links from mid-tier pages — the equity never reaches them.
32. Anchor text is descriptive, not generic. “Click here” and “read more” waste the semantic signal anchor text can send. Common mistake: navigation menus using generic text for links to important pages. Use keyword-relevant anchor text where natural.
33. No broken internal links. Broken links (404 destinations) waste PageRank and degrade user experience. Crawl the site regularly with a tool that flags broken internal links.
34. Hub pages exist for each major content cluster. A hub or index page for each content cluster collects links to all cluster articles and receives links from each cluster article in return. Common mistake: clusters with no hub — the cluster pages exist in isolation without a unifying index page Google can treat as a topical authority signal.
Bonus: Audit Items Often Missed
35. robots meta tags are correct in both HTML and HTTP headers.
An X-Robots-Tag: noindex in the HTTP header overrides a page’s HTML robots meta tag. Common mistake: CDN or proxy layers injecting noindex headers that block indexation silently.
36. Image alt text is populated and descriptive. Alt text is used by screen readers and by Google Image Search. Common mistake: all images with empty alt text, or with generic filenames (“image001.jpg”) that provide no semantic signal.
37. Page titles are unique and under 60 characters. Duplicate titles confuse Google about which page is most relevant for a query. Common mistake: CMS templates generating identical titles across category pages (e.g., “Category — Site Name” for every category).
38. Meta descriptions are unique and encourage clicks. Meta descriptions don’t directly affect rankings, but they affect CTR — which does. Common mistake: truncated or auto-generated meta descriptions that cut off mid-sentence.
39. Structured internal search is handled correctly. If your site has internal search, the search result pages should be blocked from indexation (noindex or disallowed). Common mistake: search result pages getting indexed, creating thousands of thin, dynamically generated pages.
40. Hreflang return tags are symmetric. Every page in an hreflang group must include a reference to every other page in the group — including itself. Common mistake: adding hreflang to new translated pages without updating the hreflang on the original pages.
How Muginai Automates Technical SEO Audits
Running this checklist manually against dozens of URLs is an afternoon’s work. Running it continuously — catching regressions introduced by each deployment — requires automation.
Muginai’s technical SEO audit module runs the full checklist on a scheduled cadence, compares results against the previous run, and pushes delta alerts when new issues appear. You see what broke, when it broke, and which URLs are affected — without logging into a dashboard to find out. The audit integrates with Search Console data to correlate technical issues with ranking changes, making it possible to diagnose the root cause of traffic drops within hours rather than weeks.
Technical SEO is maintenance work. Muginai treats it as infrastructure monitoring rather than a quarterly audit project. Join the waitlist to see how the audit module works.