Checklist

SEO Go-Live Checklist

Last updated 14 July 2026

Use this checklist before launching a new site or going live after a redesign. Run it on your staging environment first to catch issues before they hit production, then repeat the live checks after launch.

For site migrations specifically (domain change, URL restructure, CMS switch), see the Site Migration Guide. For pre-launch planning decisions (domain, CMS, architecture, entity signals), see the New Website SEO Guide.

Crawling and indexing

Staging site is blocked from crawling (robots.txt disallow or password protection - do not rely on robots.txt alone)
Production site has no Disallow: / in robots.txt
No unintended noindex meta tags on pages that should be indexed
No X-Robots-Tag noindex headers on indexable pages

Remove staging noindex before launch

Sites are commonly built with a noindex directive in place so internal teams can preview the build without it appearing on Google before launch. This is correct practice during development, but the directive must be removed before going live. A site that launches with noindex active will not appear in search results at all.

AI crawler access is a deliberate choice, not an inherited default. Decide separately for training crawlers (GPTBot, ClaudeBot), search crawlers that index for citation (OAI-SearchBot, PerplexityBot) and agent fetchers acting for a user in real time (ChatGPT-User, Perplexity-User)
Crawler blocking at the CDN or bot-management layer has been checked, not just robots.txt. A rule there can block a crawler your robots.txt allows, and the file will not show it. If the site is on Cloudflare and carries ads, note that from 15 September 2026 the Training and Agent categories are blocked by default on ad-displaying pages for newly onboarded domains, though Search crawlers stay allowed.¹ If you choose to block Training, be aware that also blocks Googlebot, Applebot and Bingbot¹
Canonical tags are present on all key pages and point to the correct URL (self-referencing on unique pages)
Canonical tags use absolute URLs, not relative paths
Pagination handled correctly (self-canonical, or canonical to page 1 where consolidation is intended). Note: Google dropped rel=“next”/“prev” support in 2019
No orphan pages - all key pages are reachable via internal links

XML sitemap

XML sitemap exists and is accessible at /sitemap.xml
Sitemap is declared in robots.txt
Sitemap contains only canonical, indexable URLs (no noindex pages, no redirects)
Sitemap has been submitted to Google Search Console
Sitemap has been submitted to Bing Webmaster Tools

HTTPS and security

SSL certificate is valid and not expired
Site loads correctly on HTTPS
HTTP redirects to HTTPS (301, not 302)
www and non-www resolve to a single canonical version (301 redirect)
No mixed content warnings (HTTP resources loaded on HTTPS pages)
HSTS header is set

Redirects

All old URLs redirect to their equivalent new URLs (301)
No redirect chains longer than two hops
No redirect loops
Homepage redirects are correct (if URL has changed)
Old URLs are not returning 200 (ghost pages)

On-page

Every page has a unique, descriptive <title> tag, ideally 50–60 characters. This is a display and rewrite-risk target, not a ranking rule: Google sets no length limit and truncates to the device width, but titles in this range are rewritten least often, and titles over 70 characters are rewritten almost every time²
Each <title> agrees with the page’s visible H1. Matching them is the most effective way to stop Google replacing your title²
Every page has a unique meta description (140–160 characters)
Each page has a single <h1> that matches the page topic
Heading hierarchy is logical (h1 → h2 → h3, no skipped levels)
Images have descriptive alt attributes
Images are compressed and served in next-gen formats (WebP, AVIF)
No duplicate title tags or meta descriptions across the site

Structured data

Schema markup is present on key page types (articles, products, local business, etc.). FAQPage is no longer worth shipping for rich results: Google removed the FAQ rich result in May 2026 and deleted its documentation. It retains value for AI extraction only
Structured data is valid - test with Google’s Rich Results Test
No structured data errors in Search Console

Core Web Vitals and performance

LCP (Largest Contentful Paint) is under 2.5 seconds
INP (Interaction to Next Paint) is under 200ms
CLS (Cumulative Layout Shift) is under 0.1
PageSpeed Insights score reviewed for mobile and desktop
Render-blocking resources addressed (defer/async JS, preload critical CSS)
Fonts are preloaded or system fonts used to prevent layout shift

Mobile

Key templates checked on mobile (Chrome DevTools device emulation or PageSpeed Insights). Google’s standalone Mobile-Friendly Test was retired in December 2023, and mobile-friendliness cannot be read off the Lighthouse SEO score: version 12.0.0 moved tap-targets to accessibility as target-size, and version 13 removed the font-size audit outright³
Viewport meta tag is present (Lighthouse reports this as viewport-insight from version 13³)
Text is readable without zooming. Lighthouse no longer audits this: check it by hand
Tap targets are appropriately sized and spaced (Lighthouse: accessibility, target-size)
No content is hidden behind intrusive interstitials

International (if applicable)

hreflang tags are implemented correctly on all language/region variants
hreflang uses correct ISO language and country codes
Each hreflang tag includes a reciprocal tag on the target page
x-default hreflang is set for the fallback page

Analytics and Search Console

Google Search Console property is verified for the production domain
Bing Webmaster Tools is verified
GA4 is installed and tracking pageviews correctly
GA4 goals/conversions are configured
Internal traffic is filtered (IP exclusion or developer traffic filter in GA4)
Search Console is linked to GA4
GA4 AI Assistant channel is visible (traffic from ChatGPT, Claude, Gemini, etc. will appear here automatically)

Monitoring setup

Uptime monitoring is configured (alerts on downtime)
Core Web Vitals monitoring is set up (Search Console, or a third-party tool)
Rank tracking is configured for target keywords
Search Console email alerts are enabled

Immediately after launch

Crawl the live site with Screaming Frog or Sitebulb
Confirm the sitemap is accessible and valid on the live domain
Verify robots.txt on the live domain permits crawling
Check 10–20 key pages for correct title, canonical, and indexing status using URL Inspection in Search Console
Confirm GA4 is tracking sessions on the live domain
Monitor Search Console for crawl errors in the first 72 hours
Check Core Web Vitals on the live site (staging performance can differ from production)

2–4 weeks post-launch

Key pages are appearing in search results
No significant drop in organic traffic (compare to pre-launch baseline)
Search Console Page Indexing report shows no unexpected errors or newly non-indexed pages
Core Web Vitals passing in Search Console field data
Backlinks to old URLs (if applicable) are resolving correctly via redirects