If your existing website gets traffic from Google, replacing it can feel slightly uncomfortable. The website might need redesigning, but some of what’s sitting underneath it has taken years to build.
Pages may rank for useful searches. Other websites might link to them. Google has crawled and indexed the site over time, and some of those pages may be responsible for enquiries even if nobody inside the business pays much attention to them.
The good news is that changing how a website looks doesn’t mean starting again with Google. The important part is understanding what you’re changing underneath the design and making sure anything worth keeping survives the move.
A redesign should start by finding out what the existing website has already earned.
First, find out what you have to lose
One of the easiest mistakes to make during a redesign is assuming you already know which pages matter.
The homepage obviously matters. Your main service pages probably matter. Beyond that, things can become less predictable.
An old article might be bringing in useful search traffic every month. A service page nobody particularly likes might rank well for an important search. A fairly ordinary-looking page could have links pointing to it from other websites. There may also be pages that get very little traffic but play an important part in helping visitors understand the business.
A page doesn’t have to look important to be important.
Before planning the new website, it’s worth understanding which pages receive organic traffic, what people are searching for when they find them, which pages are indexed and where useful external links are pointing. Google Search Console, analytics data, your existing sitemap and backlink information can all help build that picture.
This doesn’t mean everything performing well has to be preserved exactly as it is. It means you should know what you’re changing before you change it.
Otherwise, you’re making decisions about the new website without knowing which parts of the old one are already doing useful work.
Keep good URLs when there’s no reason to change them
A new website doesn’t need new addresses for every page.
If your accountancy firm’s tax service already lives at /business-tax/, that address still makes sense and the service still exists, there may be very little to gain from changing it simply because the rest of the website is being redesigned.
URL changes can be handled properly, but every unnecessary change creates another thing that needs handling.
This is particularly important when redesign projects become tidying exercises. Someone decides all the URLs should follow a new format, categories should be renamed or every service should move into a different folder because the new structure looks neater on a spreadsheet.
Sometimes those changes are worthwhile. If the existing structure is genuinely confusing or no longer reflects the business, changing it can make sense. But a redesign isn’t an automatic reason to rename everything.
If an existing URL is clear, relevant and fits comfortably into the new website, leaving it alone is often the simplest option.
When a URL does change, give the old one a proper destination
Sometimes URLs genuinely need to change. Services are renamed, pages are combined and the structure of the business may have changed enough that carrying the old website architecture forward would defeat the point of redesigning it.
That’s where redirects matter.
Suppose the old website has a page at /business-tax/, but the equivalent page on the new site is going to live at /services/business-tax/. Anyone following an old link, bookmark or Google result needs to be taken from the old address to the new one.
A permanent server-side redirect, usually a 301 or 308, provides that route and tells search engines that the content has permanently moved.
This should be planned rather than dealt with after launch. Create a record of the important URLs on the existing website and decide what will happen to each one. Some will stay exactly where they are. Some will move to a new address. Some may be combined with another page, while others may genuinely no longer have a useful replacement.
Google recommends mapping old URLs to their appropriate new destinations and using permanent redirects when URLs change. It also states that 301 and other permanent redirects don’t cause a loss of PageRank.
The important word in all of this is appropriate.
Don’t redirect everything to the homepage
When an old website contains dozens of pages that aren’t appearing on the new one, there’s an attractive shortcut: redirect all of them to the homepage.
Everything has a destination. No more 404 errors. Job done.
Except those destinations don’t necessarily make any sense.
If someone follows a link expecting detailed information about a particular service and suddenly arrives on a general homepage, the redirect hasn’t really helped them. Search engines can reach a similar conclusion. Google warns that redirecting lots of old URLs to an irrelevant destination such as the homepage can be treated as a soft 404.
Redirects should follow meaning rather than convenience.
If three old pages covering closely related versions of a service are being replaced by one genuinely useful new service page, redirecting all three there can make complete sense. If an old page has a direct equivalent on the new website, send it there.
If a page has disappeared and there genuinely isn’t a relevant replacement, forcing it to the homepage isn’t necessarily better than allowing it to disappear properly.
A redirect isn’t just a technical way of getting rid of an error. It’s supposed to help someone find what replaced the thing they were looking for.
A redesign is also a content migration
The word “redesign” makes the project sound visual, but established websites rarely change only their appearance.
Copy gets rewritten. Services are reorganised. Pages are merged. Navigation changes. Old articles disappear. New sections are introduced. Content that once occupied 800 words might become three sentences because the new layout looks cleaner that way.
Some of those changes can make the website substantially better. They can also change what search engines have been using to understand a page.
That doesn’t mean you should preserve poor content forever because it happens to rank. SEO shouldn’t become an excuse for keeping a website frozen in time. If the business has changed, the content should change with it.
The important thing is understanding what the existing page is doing before rewriting it.
If a service page ranks for commercially useful searches, look at the queries bringing people to it and the information it contains. Work out why the page is useful before deciding what the replacement should say. You might produce something considerably better, but you’re improving it with evidence rather than throwing it away and hoping the new version performs the same way.
This is particularly relevant to the fashion for making websites extremely minimal. Removing waffle is good. Removing information simply because a shorter page looks cleaner isn’t necessarily the same thing.
Don’t confuse less clutter with less information.
A good redesign should make useful information easier to understand, not make it disappear.
Internal links need to survive the move too
URLs aren’t only reached from Google. Pages on your website link to one another, and those relationships help both visitors and search engines understand which content is important and how different subjects fit together.
A redesign can change those relationships dramatically. Navigation gets simplified, footer links disappear, articles are reformatted and service pages move around.
If an important page previously received useful internal links from several places and the new website leaves it sitting on its own, you’ve changed more than its appearance.
When URLs change, Google recommends updating internal links so they point directly to the new addresses rather than relying on redirects. That’s sensible for visitors too. There is little reason to deliberately send someone through an old address when you already know where the new page lives.
Internal linking doesn’t need to become an elaborate SEO exercise. The basic principle is much simpler: important pages should be easy to reach, and relevant content should link together where doing so genuinely helps the reader.
When rebuilding the site, don’t forget those relationships are part of what you’re migrating.
Be very careful with the staging website
A new website will usually be built somewhere private before it replaces the existing one. That’s sensible. You don’t want customers or search engines discovering a half-built website containing unfinished copy and test pages.
One common way to deal with this is to tell search engines not to index the staging website. A noindex instruction can be entirely appropriate during development.
The problem comes when that instruction follows the website onto the live domain.
If the finished site is still telling Google not to index its pages after launch, you’ve created a much more serious problem than a slightly misplaced heading or forgotten meta description.
That’s why search visibility needs to be part of the launch process rather than something checked a few weeks later when somebody notices traffic has fallen.
Development restrictions should be verified before launch and checked again once the website is live. The fact that a staging website was correctly hidden from search engines yesterday doesn’t prove that the live website is correctly available to them today.
Check what the new pages say is the preferred version
There’s another technical detail worth knowing about during a redesign: canonical URLs.
A canonical helps search engines understand which URL is the preferred version when the same or very similar content can be reached through more than one address.
You don’t need to become an expert in canonicalisation to commission a website redesign. What matters is that the new website isn’t accidentally pointing search engines back towards an old address, temporary URL or staging version.
Once the new site is live, important pages should identify the appropriate live URLs as their preferred versions. Google also uses other signals, including redirects and sitemap inclusion, when working out which URL should be treated as canonical.
It’s a small technical detail with a fairly simple business-owner version: after moving house, make sure the website isn’t still telling Google that it lives at the old address.
Give Google a clean map of the new website
An XML sitemap is essentially a machine-readable list of the URLs you want search engines to know about.
After a redesign or migration, it should contain the final live URLs rather than staging addresses, deleted pages or URLs that immediately redirect somewhere else.
Submitting the new sitemap through Google Search Console gives Google another clear way to discover the new structure and is part of Google’s own guidance for site moves.
Again, this isn’t magic. Submitting a sitemap doesn’t guarantee rankings and it doesn’t repair a badly planned migration. It’s simply one of the signals that helps search engines understand what now exists.
The important thing is that all the signals agree. Redirects point to the right places, internal links use the new URLs, canonicals reference the correct versions and the sitemap lists the pages you actually want indexed.
Launch isn’t the end of the migration
A website can look completely finished to a human visitor while Google is still processing what changed.
That’s why the first few days and weeks after launch matter.
Google Search Console can show whether pages are being indexed and highlight crawl problems. Analytics can help identify unexpected changes in organic traffic. Important pages and search queries can be monitored to see whether visibility is behaving roughly as expected.
Google advises website owners to expect some temporary ranking fluctuation during significant site moves while it recrawls and reindexes the changed URLs. Depending on the size of the site and the extent of the changes, that process can take time.
So movement after launch doesn’t automatically mean something has gone wrong. But it does mean somebody should be watching.
Pay particular attention to sudden problems rather than obsessing over every small daily movement. Important pages disappearing from search, large numbers of unexpected 404 errors, redirects failing, pages remaining noindex or Google discovering the wrong canonical URLs all deserve investigation.
A migration isn’t finished because the new homepage looks good. It’s finished when the new website is working properly for visitors and search engines have been given a clear path from what existed before to what exists now.
An SEO plugin doesn’t protect a redesign
WordPress makes it easy to install an SEO plugin. Tools such as Rank Math and Yoast can be useful for managing titles, descriptions, schema, sitemaps and other technical settings.
What they can’t do is understand your business and plan the migration for you.
Installing an SEO plugin on a new website doesn’t tell you which old pages receive valuable traffic. It doesn’t decide whether /business-tax/ should keep its existing URL. It doesn’t know whether three old pages should become one new page or whether removing a particular piece of content will matter.
Those decisions happen before and during the redesign.
Calling a website “SEO-friendly” because an SEO plugin has been installed is therefore missing most of the point. Protecting existing search visibility starts with understanding the old website, deciding carefully what will change and making sure search engines can follow those changes.
The plugin is a tool. It isn’t the strategy.
Should you avoid redesigning a website that already ranks?
No.
Keeping an increasingly unsuitable website forever because some of its pages perform well in Google isn’t a particularly good long-term SEO strategy.
If the business has changed, the website may need to change with it. Perhaps the services are different, the positioning has moved on or the structure no longer makes sense. The existing website might rank reasonably well while doing a poor job of converting the people who arrive.
The answer isn’t to choose between a better website and your existing search visibility.
It’s to redesign the website while carrying forward the things the old one has already earned.
That means treating the existing site as an asset rather than simply something to replace.
A simple way to approach a website redesign
You don’t need to become an SEO specialist to understand the basic process.
Before° — Find out what already works. Understand which URLs receive traffic, which searches matter, where external links point and which pages are worth protecting.
Build° — Keep sensible URLs where possible. When things need to move, plan where they will go. Improve useful content rather than accidentally stripping it away, and make sure important pages remain properly connected.
Launch° — Test the handover. Check redirects, internal links, canonicals and the sitemap. Make sure development restrictions such as noindex haven’t made their way onto the live website.
Watch° — Use Search Console and analytics to see what happens next. Look for genuine crawl and indexing problems, monitor the pages that matter and give search engines time to process legitimate changes.
None of those steps are particularly glamorous, which is probably why they’re easy to overlook during a project dominated by new designs.
But if an established website has spent years building search visibility, replacing it shouldn’t mean pretending none of that history exists.
A good redesign improves what customers see while protecting what was already working underneath.