The Technical SEO Checklist We Run on Every Launch

Content and links get the attention, but a surprising amount of ranking potential is lost to boring, invisible technical problems: pages search engines can’t crawl, markup they can’t understand, and sites too slow to reward. This technical SEO checklist is the one we run before — and after — every launch, and it’s written for founders and product leads who want to know exactly what “technical SEO” covers before they sign off on a build.

None of it is glamorous. All of it compounds. A site with brilliant content but a broken sitemap will still lose to a mediocre competitor who got the fundamentals right, so treat this as the floor every project needs to clear, not a nice-to-have for later.

Why a technical SEO audit matters more than it looks

Search engines can only rank what they can find, understand and load quickly. On-page copy and backlinks get a page into contention, but crawlability, site speed, mobile-friendliness and clean markup decide whether it’s even in the race. Run a technical SEO audit before launch and you catch the expensive mistakes — a stray noindex, a duplicate URL, a sitemap full of dead links — while they’re still a five-minute fix rather than a six-week ranking recovery.

Crawlability and indexing

• A clean robots.txt that blocks nothing important and points to the sitemap.
• An accurate XML sitemap that lists only canonical, indexable URLs — no redirects, no 404s, no orphaned pages.
• One canonical URL per page, with consistent trailing-slash, www and protocol rules applied site-wide.
• No accidental noindex tags left over from staging — the classic launch-day disaster that quietly de-indexes a whole site.
• A logical site architecture, so every important page is reachable within a few clicks of the homepage rather than buried in a menu no crawler ever finds.

If you’re rebuilding or moving an existing site rather than launching a new one, crawlability and redirects deserve even more care — old URLs, earned links and rankings are all on the line. We’ve written a longer technical SEO migration guide that goes through the redirect mapping and rollback plan we use to keep traffic intact through a rebuild.

Structured data

Schema.org markup is how you turn a plain blue link into a rich result — star ratings, FAQs, breadcrumbs, event dates. At minimum we ship Organization and WebSite schema site-wide, Article schema on blog posts, FAQPage where the content genuinely answers common questions, and BreadcrumbList for navigation. Every block is valid JSON-LD, tested against Google’s Rich Results tool and Search Console’s enhancement reports, not guessed at and left unchecked.

Structured data won’t rescue thin or duplicate content, but on a page that already deserves to rank, it’s often the difference between a plain listing and one that stands out in the results — and, increasingly, between being cited in an AI-generated summary or being skipped entirely.

Speed and Core Web Vitals

Rankings and revenue both track load time. We hold pages to a performance budget and monitor real-user LCP, INP and CLS, because a fast lab score means little if real phones on real networks tell a different story. The biggest wins are usually unglamorous: shipping less JavaScript, serving properly sized and compressed images, and avoiding layout shifts caused by fonts or ads loading late.

If your Core Web Vitals scores are already a known weak point, it’s worth reading how to improve your Core Web Vitals score properly alongside this checklist, and our write-up on shaving a website from four seconds to under one for a concrete example of what the fix actually looks like in practice. Speed isn’t purely a launch-day concern — it’s worth designing for from the first architecture decision, not bolting on once a site is already slow.

Mobile-friendliness and accessibility

Google indexes and ranks the mobile version of your site, so a layout that only works on desktop is a technical SEO problem, not just a design one. Check that tap targets are large enough, text is legible without zooming, and nothing important is hidden behind hover states that don’t exist on a touchscreen.

Accessibility and technical SEO overlap more than most teams expect: semantic HTML, meaningful alt text and a logical heading order help screen readers and search engine crawlers in exactly the same way. If accessibility hasn’t had a proper look yet, our guide to what makes a website accessible in practice is a useful companion piece to run alongside this checklist.

The fundamentals that still matter

• Semantic HTML with a single, descriptive <h1> per page.
• Descriptive, unique title tags and meta descriptions written for humans first, keywords second.
• Descriptive alt text on meaningful images, and none on purely decorative ones.
• Internal links that give crawlers — and readers — a clear map of what matters most on the site.
• HTTPS everywhere, with old URLs 301-redirected so you never throw away earned authority.
• A favicon, a working 404 page, and consistent metadata across social sharing cards — small things, but they’re what a real user or a suspicious crawler checks first.

Common mistakes this checklist catches

Most technical SEO damage isn’t caused by ignorance — it’s caused by good work being undone by a small, unglamorous oversight nobody flagged before launch. The recurring ones we see:

• Staging settings shipping to production. A password-protected staging site is often set to block all crawlers by default; if that setting survives the go-live, the entire site can sit invisible to Google for weeks before anyone notices.
• Redirect chains instead of clean redirects. Page A redirects to B, which redirects to C — each hop leaks a little authority and slows the page down. One clean 301 straight to the final URL is always better.
• Duplicate content from URL parameters. Tracking parameters, filters and session IDs can quietly generate hundreds of near-identical URLs that dilute a canonical page’s authority.
• Images with no dimensions set. Even a fast-loading image causes a layout shift, and a poor CLS score, if the browser doesn’t know its size before it renders.
• Metadata copied and never rewritten. Title tags and meta descriptions cloned from a template or a previous site tell search engines — and searchers — nothing distinctive about the page.

Each of these is a five-minute fix if it’s caught before launch, and a much longer one if it’s discovered three months later in a traffic report.

Running it as a repeatable process

The checklist only earns its keep if it’s run the same way every time, not improvised under launch-day pressure. We treat it as a standard step in every build we ship, alongside performance testing and a final content review, so nothing gets waved through because the deadline is close. If you want to see how this fits into a full engagement rather than a one-off audit, our services overview covers where technical SEO sits alongside the rest of the build.

None of this is glamorous. All of it compounds. Get the technical foundation right and every piece of content you publish afterwards works harder — instead of quietly leaking rankings to problems nobody thought to check.

If you’d like a second pair of eyes on your site before or after launch, book a free consultation and we’ll walk through exactly where the opportunities are.