Relaunching a website without losing visibility: redirects, sitemaps and what Google says
In a relaunch, addresses, structure and technology often change at the same time. If old addresses are not cleanly redirected to new ones, links, bookmarks and part of the site's visibility in search engines are lost. This article explains the technical basics, summarises Google's recommendations and contains a step-by-step plan.
by Kilian Dickmann, Founder and ownerPublished on 6 min read
For many small businesses, a relaunch is the opportunity to replace a website that has aged. New design, new copy, often a new system as well. What is easily overlooked: over the years, the old website has built up addresses that are listed in search engines, directories, on business cards and in bookmarks. If these addresses are simply abandoned, visitors land on error pages and search engines have to get to know the website partly from scratch.
What happens in a relaunch
Google distinguishes between a move without URL changes, for example when changing hosting, and a move with URL changes [1]. The latter is the normal case in a relaunch: /services.php?id=3 becomes /services/heating, http:// becomes https://, pages are merged or dropped.
For every old address, the question is then what a visitor or a search engine crawler finds there. The answer is given by the HTTP status code the server returns.
The status codes
| Code | Meaning | Use |
|---|---|---|
| 301, 308 | Moved permanently | The page has a new address. The standard case in a relaunch. |
| 302, 307 | Moved temporarily | Only if the old address will apply again later, for example during maintenance. |
| 404 | Not found | The page does not exist and has no equivalent. |
| 410 | Gone | The page was deliberately removed. |
The semantics of these codes are defined in the HTTP standard [8]. Google treats permanent redirects as a strong signal that the new address should appear in search results in future, temporary ones as a weaker signal [2]. Redirects should be set up server-side where possible; Google does recognise meta refresh and JavaScript redirects but only recommends them if server-side redirects are not possible [2].
A common concern is that redirects cost “link power”. Google’s documentation on site moves states explicitly that 301 and other permanent redirects do not cause a loss in PageRank [1].
The redirect map
The most important tool in a relaunch is a table that maps every old address to a new one. It is created in three steps:
- Capture all old addresses. Sources are the old sitemap, a list of all pages from the previous system, the performance reports in Search Console and, if available, access statistics. Images and PDF files linked from outside belong to it as well.
- Assign each address a destination with matching content. The old page on heating maintenance redirects to the new page on heating maintenance, not to the homepage.
- Handle pages without an equivalent deliberately. If there is no longer any matching content, a 404 or 410 is more honest than a redirect to some page.
The third point matters: if many old addresses are redirected wholesale to the homepage, Google may treat this like an error page, because the destination does not match the original in content [2] [6].
Avoid chains and loops
If A redirects to B and B to C, a chain is created. Google follows a limited number of redirects; every step costs time and increases the risk of errors [2]. After a relaunch, every old address should reach its destination in one step. This particularly affects websites that have been through earlier moves, for example from http to https or from an address without www to one with it.
Preparing the new website
Canonical addresses
Every page should be reachable at exactly one address. Variants with and without www, with and without a trailing slash or with parameters are consolidated into the preferred version by redirect or via rel="canonical" [5]. Internal links point directly to this version, not to addresses that are only redirected.
Sitemap
An XML sitemap lists the addresses that should be indexed and can be submitted in Search Console [3]. For the relaunch, it contains the new addresses. It speeds up discovery but does not replace redirects.
Multiple languages
If there are language versions, they refer to each other via hreflang, each with full addresses and reciprocally [7]. If the addresses change in the relaunch, these references must be updated too.
No blocks left over from testing
A common and consequential mistake: during development, the new website was hidden from search engines with noindex or a block in robots.txt, and this block remains in place at launch [9]. Before switching over, check specifically that the production pages can be indexed and test environments cannot.
The process
- Inventory of the old addresses and their importance (traffic, external links).
- Create the redirect map and have it checked for content: does every destination fit?
- Prepare the new website with canonical addresses, a sitemap, titles and descriptions.
- Test the redirects before they go live, ideally automatically across the whole list: every old address must lead in one step with 301 or 308 to a destination with status 200.
- Switch over, ideally during a quiet period. Google recommends splitting large changes into steps and not changing everything at once, where that is possible [1].
- Use Search Console: submit the sitemap, test important pages with URL inspection. When changing the domain, also use the Change of Address tool; it is not intended for a move from
httptohttps[4]. - Monitor: follow error reports, pages not found and search performance over several weeks. Google points out that rankings may fluctuate temporarily during a move [1].
- Keep redirects permanently. For domain changes, Google recommends maintaining redirects for at least 180 days, and longer as long as traffic still arrives from search [4]. Since external links and bookmarks often persist for years, it makes sense in practice not to remove them.
Common mistakes
| Mistake | Consequence |
|---|---|
| No redirects | Old links lead to error pages; visibility has to be rebuilt |
| Everything to the homepage | May be treated like an error page [2] |
| Temporary instead of permanent redirect | Weaker signal for the new address [2] |
| Redirect chains | Slower, more error-prone [2] |
| noindex left over from development | New pages do not appear in search [9] |
| Internal links to old addresses | Unnecessary redirects, inconsistent signals [5] |
| Redirects removed after a short time | Links found later lead nowhere [4] |
Conclusion
A relaunch need not cost visibility if it is treated as a move rather than a fresh start. The effort lies less in the technology than in the care taken: capture every old address, give it a destination with matching content, redirect permanently, test, and monitor afterwards. Google describes this route in detail in its documentation; following it gives search engines and visitors the chance to find the new website just as they found the old one.
Frequently asked questions
- Do 301 redirects cost rankings?
- According to Google's documentation, 301 and other permanent redirects do not cause a loss in PageRank. Temporary fluctuations during a move are still possible.
- May I redirect all old pages to the homepage?
- Better not. Google may treat mass redirects to destinations that do not match in content like error pages. Every old address should point to the matching new page.
- How long do redirects have to stay in place?
- For domain changes, Google recommends at least 180 days, and longer as long as traffic still comes from search. In practice it makes sense to keep them permanently.
Sources
The numbers in square brackets in the text refer to this list. Online sources last checked in November 2026.
- [1]Google Search Central: Site moves with URL changes. developers.google.com/search/docs/crawling-indexing/site-move-with-url-changes
- [2]Google Search Central: Redirects and Google Search. developers.google.com/search/docs/crawling-indexing/301-redirects
- [3]Google Search Central: Sitemaps overview. developers.google.com/search/docs/crawling-indexing/sitemaps/overview
- [4]Search Console Help: Change of Address tool. support.google.com/webmasters/answer/9370220
- [5]Google Search Central: How to specify a canonical URL with rel="canonical" and other methods. developers.google.com/search/docs/crawling-indexing/consolidate-duplicate-urls
- [6]Google Search Central: How HTTP status codes, and network and DNS errors affect Google Search. developers.google.com/search/docs/crawling-indexing/http-network-errors
- [7]Google Search Central: Tell Google about localized versions of your page (hreflang). developers.google.com/search/docs/specialty/international/localized-versions
- [8]Fielding, R., Nottingham, M. & Reschke, J. (Eds.) (2022). RFC 9110: HTTP Semantics, section 15.4 Redirection 3xx. IETF. www.rfc-editor.org/rfc/rfc9110
- [9]Google Search Central: Robots meta tag, data-nosnippet, and X-Robots-Tag specifications. developers.google.com/search/docs/crawling-indexing/robots-meta-tag
About the author
Kilian Dickmann, Founder and owner
Designs and builds websites for small businesses and writes here about design, technology, accessibility and data protection in practice.