Skip to content Skip to footer

New Website SEO Checklist: Launch Day and the First Month

Summarise this article with
Browser window launching upward beside a clipboard checklist and a redirect diagram from an old page to a new page

This is a launch checklist, not a general SEO guide. It covers the fortnight before a new website goes live, the launch day itself and the first month afterwards, which is the window where most of the avoidable damage happens.

Why a new website needs its own checklist

A brand new site and a rebuild of an existing site are different problems, and most new website SEO checklists quietly assume the first while the majority of readers are doing the second. If you are replacing a site that already gets traffic, your priority is not to gain rankings, it is not to lose the ones you have, and that is a redirect and continuity problem before it is anything else. Our guide to website migration SEO covers that case in full, and you should work through it alongside this list rather than instead of it.

If the site is genuinely new, with no predecessor and no existing rankings, the risk is different: launching with the settings that were sensible on staging and never changing them. Both failure modes are entirely preventable and both are extremely common.

The six things to get right at launch

These are the items where getting it wrong on launch day costs weeks to undo.

Indexability switched on

The staging noindex and the robots.txt disallow both need to come off, and both need to be verified on the live domain rather than assumed. This is the most common launch day failure there is.

A complete redirect map

Every old URL that had traffic, a link or a ranking needs a one hop redirect to its closest equivalent. Not to the homepage. Mapped before launch, tested after.

Analytics and Search Console live

Property verified, tracking installed and firing on the live domain, conversions defined. Without this you cannot tell whether launch went well, only whether it felt like it did.

Titles and descriptions written

Not generated from the template, not left as the theme default, and not duplicated across the whole site. Every page that matters gets its own.

One canonical version of the site

One protocol, one host, one trailing slash convention, everything else redirecting to it in a single hop. Settle this before launch, not after links start accumulating.

A sitemap submitted on day one

Generated from live canonical URLs, listed in robots.txt and submitted to Search Console. It is the fastest signal you can send that the site exists.

Before launch

Do all of this while the site is still on staging. Every item here is cheaper now than it will be in a week.

  • Export the full URL list from the old site, with the last twelve months of traffic, impressions and inbound links against each.
  • Build the redirect map from that export. Every URL with traffic, a ranking or a link gets a specific destination. Anything genuinely obsolete can go to the most relevant category page, but blanket redirects to the homepage are treated as soft 404s and lose the value entirely.
  • Match the new URL structure to how the site is organised, then stop changing it. Every later change is another redirect.
  • Write unique title tags and meta descriptions for every page that matters. Templates are fine for large catalogues, defaults are not.
  • Confirm the heading hierarchy is sane on the templates: one H1 per page, H2 for sections, no skipping levels for styling reasons.
  • Compress and correctly size every image in the theme and the content. This is almost always the biggest speed win available before launch.
  • Set up the structured data the site actually needs: Organisation, plus Article, Product or LocalBusiness as appropriate.
  • Check every internal link in the new content points at the new URL structure, not at old paths that will immediately redirect.
  • Keep the old site accessible somewhere you can reference it. You will want to check something.

Launch day

Work through these in order, in the first hour after the site goes live. Do not leave them to the end of the day.

  1. Remove the noindex. Check the live page source for a robots meta tag and check the SEO plugin's site wide setting. Both. They are separate switches and either one will keep you out of the index.
  2. Check robots.txt on the live domain. A staging disallow that gets deployed with the site is the classic version of this failure.
  3. Spot check twenty redirects by hand. Pick your highest traffic old URLs and confirm each returns a single 301 to the right page, with no chain and no loop.
  4. Confirm HTTPS across the whole site. Including images, scripts and fonts. Mixed content warnings on launch day are avoidable.
  5. Verify analytics is firing on the live domain, in a private window, and that a test conversion is recorded.
  6. Submit the sitemap in Search Console and request indexing for the homepage and your most important pages.
  7. Check the site on a real phone on mobile data. Not the responsive preview in a desktop browser.

The first month

The work after launch is monitoring, and it is the part people skip because the project feels finished. It is not finished for about four weeks.

  • Check Search Console coverage weekly. Newly excluded pages are the early warning for a redirect or canonical problem.
  • Watch for a crawl error spike. A small rise is normal as old URLs are re-crawled; a sustained one means the redirect map has gaps.
  • Compare impressions and positions for your top twenty queries against the pre-launch baseline. A modest dip for two to three weeks is normal on a rebuild. A cliff is not.
  • Find and fix any redirect chains that emerged once the old rules met the new ones.
  • Update your Google Business Profile, social profiles and any directory listings with the new URLs, especially if the structure changed.
  • Re-run a crawl at the end of week four and compare it against the pre-launch crawl.

Building the redirect map in practice

This is the part that decides whether a rebuild keeps its traffic, so it is worth doing properly rather than from memory. Start with three exports and combine them: every URL Search Console has served an impression for in the last twelve months, every URL in your analytics with a session in the same window, and a full crawl of the old site. The three overlap heavily but none of them is complete on its own. Search Console misses pages that never ranked but carry links, analytics misses pages nobody visited, and a crawl misses orphan pages that are still linked from elsewhere on the internet.

Then sort the combined list by value rather than alphabetically, and work down. The top of that list, typically a few dozen URLs on a small business site, deserves an individually chosen destination. Below that, pattern rules are fine: an old blog path mapping to a new blog path, a service category mapping to its replacement. At the bottom sit genuinely dead pages, and those can go to the nearest relevant parent, which is not the same thing as the homepage.

Two checks catch most errors. First, no destination should itself be a redirect, or you have built a chain on day one. Second, every destination should return a 200 on the new site, which sounds obvious and is the single most common defect in a hand built map, because destinations get renamed during the build after the map was written.

A two week countdown

If you want the whole thing as a schedule rather than a list, this is the shape that works for a typical small business rebuild.

  • Two weeks out. Pull the three exports, build the redirect map, freeze the new URL structure. Take the ranking and traffic baseline and save it somewhere you will find it again.
  • One week out. Titles, descriptions and heading structure finalised. Images compressed. Structured data in place on staging. Crawl the staging site and fix what the crawl finds.
  • Two days out. Load the redirect rules and test them against staging. Confirm analytics and Search Console are ready for the live domain.
  • Launch day. The seven checks above, within the first hour.
  • Day two. Full crawl of the live site. Fix any chain, loop or 404 it surfaces.
  • Weeks one to four. Weekly coverage and crawl error review, then a comparison against the baseline at the end of week four.

What people get wrong

Three mistakes cause most post launch traffic losses, and none of them are subtle once you know to look.

The first is redirecting everything to the homepage. It is fast, it makes the 404 report look clean, and it throws away the accumulated value of every page you redirected. A redirect only passes value when the destination is a genuine equivalent of the source.

The second is launching without a baseline. If you did not record positions, impressions and conversions before launch, you cannot answer the only question that matters afterwards, which is whether this went well. Take the export before you start, not after somebody asks.

The third is treating the first fortnight's numbers as the verdict. Re-crawling takes time, and a rebuild almost always dips before it recovers. Judging at day ten produces panic changes that make the recovery slower, not faster.

There is a fourth mistake that is less dramatic and more expensive over time: nobody owns the site after launch. The agency's involvement ends at go live, the internal owner assumes it is done, and the four weeks of monitoring that would have caught the redirect gap simply do not happen. Agree before launch who is checking coverage in week two and who fixes what they find. It takes an hour a week and it is the difference between a dip and a loss.

New site with no predecessor

If there is genuinely nothing to migrate, the list shortens considerably. You still need the indexability checks, the canonical decision, the titles and descriptions, the sitemap and the analytics. You do not need the redirect map, and you should set expectations accordingly: a new domain with no history takes months rather than weeks to establish itself, and no amount of launch day diligence changes that. What launch day diligence does is make sure that when the site does start to earn rankings, nothing technical is standing in the way.

From there the work becomes ordinary optimisation. Our technical SEO checklist is the quarterly maintenance list, the local SEO checklist applies if you serve a specific area, and Google's own search essentials starter guide is a reasonable non commercial reference for the fundamentals. If you would rather have it handled, that is what our SEO service does.

FAQ

What should be on a new website SEO checklist?

Three phases. Before launch: a URL export and redirect map, a settled URL structure, unique titles and descriptions, compressed images and structured data. Launch day: remove the staging noindex, check robots.txt on the live domain, spot check redirects, confirm HTTPS, verify analytics and submit the sitemap. First month: weekly coverage checks, crawl error monitoring and a comparison against the pre-launch baseline.

Will I lose rankings when I launch a new website?

On a rebuild, expect a dip of two to three weeks while search engines re-crawl and re-evaluate the new URLs. That is normal and it recovers. A sustained drop that does not begin to recover after a month usually means the redirect map has gaps, pages are redirecting to the homepage rather than to equivalents, or a staging noindex directive is still live. All three are findable and fixable.

How long before a new website appears in Google?

A few days to a couple of weeks for the site to be indexed, assuming it is crawlable and you have submitted a sitemap. Ranking competitively is a different question and takes months on a new domain with no history. Indexation is a technical outcome you control; ranking is not.

Do I need to redirect old URLs if the new site has better content?

Yes. Redirects preserve the accumulated value of inbound links and existing rankings, and that value is attached to the old URL regardless of how good the replacement page is. Skipping the redirect map means starting from zero on every page that previously earned anything, which is the most expensive avoidable mistake in a rebuild.

What is the most common new website SEO mistake?

Launching with the staging noindex or robots.txt disallow still in place. It is a ten second check, it produces no visible symptom on the site itself, and it can cost weeks before anybody notices that nothing is being indexed. Check both the page level robots meta tag and the SEO plugin's site wide setting, because they are separate switches.

Should I keep the same URLs on a rebuild?

If the existing URLs are sensible, yes. Keeping them removes the largest single source of risk in a rebuild. Change them only where the current structure is genuinely confusing or blocks the new site architecture, and where you change them, map every one to a specific equivalent rather than to a category page or the homepage.

Aidan van Vuuren

About the author

Aidan van Vuuren

Aidan van Vuuren is the Head of Digital at Peak Digital, a UK digital marketing agency working with clients across sectors on web design, SEO, and paid media.

Connect on LinkedIn

Want help with your own marketing?

Get a free audit, we will tell you what is working, what is not, and what to do about it.

Get a free audit