Website QA Checklist Before Launch: The Technical Pre-Flight

Everyone has a launch checklist. Most of them cover the obvious things: does the site load, does the form submit, does the login work. The list below is the other list - the one for the things that break silently, cost you rankings, and come back to haunt you three months after launch when someone notices.

It's organized by category. Start with links and assets - a crawler covers most of these automatically. Then work through the manual checks that require a human to verify.

Links and navigation

Run a full site crawl and verify zero broken internal links (4xx/5xx responses on internal URLs).
Verify external links return 2xx or acceptable redirects - not 404s or 5xx errors.
Check anchor links in documentation, long-form content, and tables of contents.
Confirm all old URLs that changed redirect (301) to their new equivalents.
Verify no redirect chains of 3+ hops exist on internal links.
Test navigation on mobile - tap targets, dropdown behavior, hamburger menu.
Confirm footer links all resolve correctly.
Verify breadcrumb links are correct on all page types.

Images and assets

Verify no broken image URLs (src returning non-2xx).
Check for missing alt text on content images (not decorative images).
Confirm all JavaScript and CSS files load without errors.
Check for mixed content warnings - HTTP assets on HTTPS pages.
Verify favicon loads correctly in the browser tab and on mobile home screen.
Confirm Open Graph images are accessible and the correct dimensions (1200x630 for OG, 800x418 for Twitter card).

SEO and metadata

Every page has a unique, descriptive <title> tag (50-60 characters).
Every page has a unique meta description (150-160 characters).
Every page has exactly one <h1>.
Heading hierarchy is logical - no skipped levels (H1 to H3 with no H2).
Canonical tags are present and correct on all pages. Self-referencing on unique pages; pointing to the canonical version on paginated or filtered pages.
robots.txt exists and doesn't accidentally block important content (check for Disallow: / left over from a staging config).
XML sitemap exists, is submitted to Google Search Console, and doesn't include pages you don't want indexed.
The <html lang> attribute is set correctly.
Viewport meta tag is present for correct mobile rendering.

The robots.txt gotcha. Staging environments often have Disallow: / in robots.txt to prevent indexing during development. If you copy the staging config to production, you've just asked Google not to index your site. Check this one manually, every time.

SSL and security

SSL certificate is valid, not expired, and covers all subdomains you're using (www and non-www).
HTTP automatically redirects to HTTPS.
www and non-www both work and redirect to the canonical version.
HTTP security headers are in place: Content-Security-Policy, X-Frame-Options, X-Content-Type-Options.
No sensitive data exposed in page source, JS bundles, or network responses (API keys, passwords, internal URLs).

Forms and functionality

Contact form submits successfully and triggers the correct notification.
Form validation works on required fields and incorrect input formats.
Error states are handled gracefully - no blank white pages or raw error messages on form failures.
If there's a search function, verify it returns results and handles empty queries.
Newsletter signup works end-to-end (form submits, confirmation email arrives).

Performance

Run a PageSpeed Insights test on the homepage and a representative interior page.
Images are appropriately sized and use modern formats (WebP or AVIF) where supported.
No render-blocking scripts in the <head> that could be deferred.
Core Web Vitals (LCP, CLS, FID/INP) are within acceptable ranges on mobile.

What a crawler covers automatically

A significant portion of this checklist - broken links, missing images, anchor errors, redirect chains, mixed content, SSL errors, missing titles and meta descriptions, missing H1s, and missing alt text - can be verified automatically with a single crawl. Running 4F before launch gives you coverage on all of these in one pass, so you can spend your manual testing time on the things that actually require a human.

Try it now - it's free, no account needed

Up to 1,000 pages. Results in minutes.

After launch

The checklist doesn't end at launch. Set up scheduled weekly scans to catch issues that appear after go-live: external links that break over time, content edits that introduce broken internal links, SSL certificates that expire. A site that passed QA on launch day can develop problems within weeks if no one is watching.