For many businesses a website redesign is a long-awaited project: a new look, a clearer structure and better content management. It is also a moment when search visibility can vanish overnight. When addresses change and the old ones are not redirected to the new ones, search engines do not know where the value of the old page went, and external links land on an error page.
This guide explains how to carry out a redesign so that old links and visibility are preserved.
Why are old addresses valuable?
An old address can have three kinds of value:
- Search visibility: the page has ranked for certain search terms.
- External links: other websites, news articles, theses and publications link to it. Links are still one of the most important trust signals for search engines.
- Direct traffic: bookmarks, newsletters, printed brochures and QR codes.
External links are often long-lived. Links in the reference lists of theses and reports, for example, stay in publication archives for decades, even if the original website has been redesigned many times. Every address that is not redirected wastes that value.
The redirect map: old address → new address
A redirect map is a table that defines what happens to every old address. The map is drawn up before launch.
1. Collect the old addresses
Compile a list of all old addresses from several sources:
- the old site's sitemap and a crawl of the site, for example with Screaming Frog
- the Performance and Pages reports in Search Console
- landing pages in analytics for the last 12 months
- the list of pages with external links from link analysis tools such as Ahrefs.
The last item is important, because external links can also point to files and addresses that can no longer be found in the site's own navigation.
2. Decide on an action for each address
| Situation | Action |
|---|---|
| The content is on the new site at a different address | 301 to the matching new page |
| The content has been merged into a broader page | 301 to the broader page |
| The content has been removed, but a closely related page exists | 301 to the related page, if it genuinely meets the visitor's need |
| The content has been removed and there is no equivalent | 410 Gone |
| The address stays the same | No action, check that it works |
3. Avoid redirecting to the home page
A common shortcut is to redirect all old addresses to the home page. Google often treats such redirects as soft 404 errors, because the visitor does not find the content they were looking for. The value of the links can then be lost too. The redirect target must match the topic of the original content.
An example: an old PDF publication on funding sources for SMEs should be redirected to a new page about public funding for SMEs. That is exactly what has been done on this website: a reference that pointed to a PDF file during the domain's earlier use now redirects to the article on public funding for website and digital projects.
Implementing redirects
Redirects are implemented at server level, which makes them fast and reliable. On an Apache server this is done in the .htaccess file:
RewriteEngine On
RewriteRule ^old-service/?$ https://example.com/services/new-service/ [R=301,L]
RewriteRule ^news/2014/ - [G]
The first rule redirects a single page permanently, the second returns a 410 response for an old news archive. On an Nginx server the same is done in the server configuration file. In WordPress, redirects can also be managed with a plugin, but large numbers of redirects are best handled at server level.
Things to watch in the implementation:
- No redirect chains: the old address redirects directly to the final target, not via two or three intermediate steps.
- Query parameters: addresses in older systems often contain parameters (
?id=123). Decide whether they are carried over to the new address or dropped. - Letter case and trailing slash:
/Servicesand/services/are different addresses to the server. - Subdomains and www: addresses on old subdomains also need a DNS record and a redirect, otherwise links pointing to them break.
Launch day
- Take a backup of the old site and keep it.
- Publish the new site and switch on the redirects at the same time.
- Test the addresses in the redirect map automatically, for example with
curl -Ior a crawler, and make sure each one returns the right status code and target. - Submit the new sitemap to Search Console.
- If the domain changes too, use the Change of Address tool in Search Console.
Monitoring after launch
Search visibility usually fluctuates for a few weeks after a redesign, even when the redirects have been done well. Monitor weekly:
- 404 errors and soft 404 errors in the Pages report in Search Console
- rankings and clicks for the most important search terms
- addresses in the server logs that receive requests but are missing from the redirect map.
Add missing addresses to the redirect map as you find them. Monitoring technical SEO is covered in more detail in Technical SEO.
Frequently asked questions
What is a 301 redirect?
301 is an HTTP status code that tells browsers and search engines that a page has moved permanently to a new address. The search engine transfers the signals of the old address, such as external links, to the new page. Codes 302 or 307 are used for temporary moves.
How long should 301 redirects stay in place?
According to Google's guidance, at least a year, but in practice redirects should stay in place for as long as the old addresses receive links or traffic. External links, for example in theses and PDF publications, often survive for decades.
What is the difference between a 404 and a 410 response?
404 means the page cannot be found. 410 means the content has been removed deliberately and will not come back. Both remove the page from the search index over time. 410 states the situation more unambiguously.
Can links to an old domain benefit a new website?
Yes, if the redirect target matches the topic of the original content. If a link pointed to a funding guide, for example, and it is redirected to a completely different topic, search engines may ignore the value of the link.