Once a business decides it needs a new website, there’s a temptation to start with the exciting bits.
People collect examples of websites they like. Someone suggests changing the navigation. A conversation starts about colours, photography or whether the homepage should have a video. Before long, everyone is discussing what the new website might look like.
There’s nothing wrong with any of that. It’s just slightly early.
Before deciding what the new website should look like, it’s worth understanding why the old one is being replaced, what has changed in the business and which parts of the existing website are already doing useful work.
Most website redesign checklists start with the new website.
A better place to start is with the business it needs to represent.
1. Why are you redesigning the website?
“The website looks dated” is a reason to consider a redesign, but it isn’t much of a brief.
What has the dated website actually started causing?
Perhaps the business has moved upmarket but the website still makes it look small. Services have changed and the navigation no longer makes sense. Good prospects visit after a referral and don’t see the business they were expecting. The wrong enquiries keep coming in. Nobody on the team particularly wants to send prospective customers to the website anymore.
Or perhaps the website still looks perfectly respectable, but changing anything has become difficult enough that it is holding the business back.
Those are much more useful starting points.
Before discussing design, try finishing this sentence:
We’re replacing the website because…
The answer should describe a business problem rather than a design preference.
If the answer is difficult to articulate, spend more time there. Otherwise, six months later you may have a considerably nicer website without knowing whether you’ve solved the reason for replacing it.
2. Work out what’s changed in the business
Established businesses rarely stay still while their websites do.
The company that originally commissioned the current site may have had fewer employees, different services and a different type of customer. It may have been competing primarily on price and now compete on expertise. Perhaps most work used to be local and the business now operates nationally.
Sometimes the change is more subtle. The business hasn’t transformed, but it has become more confident about what it’s good at and which work it wants more of.
That matters because the old website was built around a version of the business.
Before rebuilding it, ask whether that version still exists.
Look at the customers you most want to attract now. Consider the services you want to grow rather than simply the ones you’ve always offered. Think about positioning, sectors, geographic reach, team expertise, pricing, how people buy and what the business wants to become known for.
Also consider what you want less of.
A website that continues to heavily promote a low-value service the business is trying to move away from isn’t simply outdated. It’s actively pulling in the wrong direction.
A redesign gives you an opportunity to align the website with where the business is going rather than where it happened to be when the previous site was built.
3. Find out what the existing website is already doing well
Replacing a website doesn’t mean everything about the existing one is bad.
Some pages may rank well in Google. Others may receive links from useful external websites. A particular article could attract a steady stream of relevant visitors. A service page might generate enquiries even though nobody inside the business particularly likes how it looks.
Before removing or changing those things, find out what they’re doing.
Look at organic traffic, Google Search Console data, analytics, existing rankings where you have reliable data, backlinks and the pages people actually visit. Speak to the team too. They may regularly send a particular page to prospects or use a resource in the sales process.
A page doesn’t have to look important to be important.
The purpose of this exercise isn’t to make the existing website untouchable. Something can perform reasonably well and still deserve improvement.
You simply want to know what you’re dealing with before you start taking it apart.
If an established website already has search visibility, changing URLs and content without understanding what was there before can create completely avoidable problems. A redesign should carry useful assets forward deliberately rather than discovering their value after they’ve disappeared.
4. Decide what you’re keeping, improving, combining and removing
“Content audit” sounds like the sort of task that produces an enormous spreadsheet nobody wants to open again.
It can be much simpler.
Take the meaningful pages on the existing website and make one of four decisions.
Keep° — The page is useful, accurate and still belongs on the new website.
Improve° — The subject still matters, but the content, positioning or presentation needs work.
Combine° — Several weak or overlapping pages would be more useful as one stronger page.
Remove° — The page no longer serves the business or its customers and doesn’t have a good reason to survive.
This forces a decision about the purpose of existing content rather than automatically copying everything into the new website.
It also prevents the opposite problem: treating a redesign as permission to delete years of content simply because the new site is supposed to be cleaner.
Some old pages deserve to disappear.
Some deserve protecting.
Most need an actual decision.
5. Build the sitemap around today’s business
One of the easiest ways to plan a new website is to open the old navigation and start moving things around.
Home stays Home. About stays About. Services gets a slightly nicer dropdown. Everything else finds roughly the same place it occupied before.
The result is a redesigned website built on the assumptions of the old one.
Instead, work backwards from what a prospective customer needs to understand.
What does the business do? Who is it particularly good at helping? What problems does it solve? Which services need proper explanation? What evidence will make those claims credible? What questions might prevent somebody from enquiring? What should they do next?
Then decide which pages are needed to answer those questions.
The existing sitemap is evidence of how the old website was organised. It isn’t a specification for the new one.
Sometimes you’ll end up with a similar structure because it was already sensible. That’s fine.
The difference is that you’ve chosen it again rather than inherited it.
6. Work out the content before design gets too far
Placeholder copy makes website design look wonderfully easy.
A heading goes here. Three lines of text go underneath. There will be a testimonial over there and perhaps a photograph next to it.
Then the real content arrives.
The heading needs two lines. There are four services rather than three. The testimonial is 90 words long. There’s an important explanation nobody considered during design. The team has six people who need showing rather than the three in the mock-up.
Suddenly the content is being forced into a design that was created before anyone knew what the page needed to communicate.
You don’t necessarily need every final sentence written before design begins. But you should understand the content properly.
Know what pages are required, what each important page needs to achieve, which messages matter, what evidence exists, what photography is available and what information still needs gathering.
Also decide who is responsible for producing and approving it.
Design and content should shape each other.
A website isn’t a collection of empty boxes that copy gets poured into afterwards.
7. Decide what will make people believe you
Established businesses often have more evidence than their websites suggest.
Years of experience. Interesting projects. Client relationships. Accreditations. Qualifications. Reviews. Results. Original photography. Recognisable customers. Specialist knowledge. Processes developed over time. People with genuine expertise.
The problem is that much of it never makes its way onto the website.
Instead, businesses write that they’re experienced, trusted and committed to quality, then leave the visitor to take their word for it.
A redesign is a good opportunity to ask a better question:
What do we know to be true about this business that would make a prospective customer more confident?
The answer will differ from one company to another.
It might be a detailed case study. It might be showing the actual people delivering the work. It could be an accreditation that genuinely matters in the industry, a customer story or photographs of completed projects.
You don’t need to turn the website into a trophy cabinet.
You need enough evidence to support the claims you’re making.
And importantly, don’t invent proof because the new design has a space labelled “statistics”.
If you don’t have a meaningful number, don’t make one up.
8. Sort out access before you urgently need it
Website projects have a remarkable ability to uncover accounts nobody knew existed.
The domain is registered somewhere, but nobody remembers where. Google Analytics was set up by a previous agency. The person with Search Console access left two years ago. The logo exists as a tiny PNG but nobody can find the original artwork.
None of these problems are particularly exciting.
They’re considerably less exciting at 4pm on launch day.
Find out who controls the domain and DNS. Identify where the current website is hosted and who has access. Check who controls analytics and Google Search Console. Understand where business email is hosted so a website move doesn’t accidentally interfere with it.
Find the original brand assets and establish ownership of photography and other important material you’re planning to reuse.
If forms connect to another system, work out what it is and who controls it.
You don’t need to migrate or change all these things. You need to know where they are and who can make decisions about them.
Do that early.
9. Decide who gets to decide
Website projects naturally attract opinions.
The managing director likes one homepage. Marketing prefers another. Sales wants more detail. Somebody thinks there should be a video. Somebody else’s partner doesn’t like the colour.
Feedback can be useful. Undefined decision-making isn’t.
Before the project gets too far, establish who owns it internally.
Who can approve the strategy? Who provides factual information? Who signs off copy? Who needs to approve the design? Who has authority to say the website is ready to launch?
That doesn’t mean one person should work in isolation.
It means the project needs a route to a decision.
Ten people can contribute useful perspectives. Requiring ten people to agree about every sentence is a very different thing.
A clear approval process protects the project from gradually becoming a collection of compromises made to satisfy whoever commented last.
10. Plan the SEO migration before launch week
If the existing website already appears in Google, search migration should be considered while the new site is being planned, not added to the launch checklist at the end.
Start with URLs.
If a useful existing URL still makes sense, there may be no reason to change it. Where URLs do need to change, decide which new page is the appropriate replacement and prepare permanent redirects.
Don’t simply send every deleted page to the homepage.
Useful content should also be reviewed before it is rewritten or removed. Existing internal links need updating where addresses change. The new site’s canonical URLs should point to the intended live versions, and its XML sitemap should contain the final URLs you actually want search engines to discover.
Google recommends this kind of deliberate mapping and permanent redirection when URLs change during a site move.
The important principle is straightforward:
Don’t wait until the old website has disappeared to work out what needed carrying across.
If your existing website receives meaningful organic traffic, this deserves its own migration plan rather than a single checkbox labelled “SEO”.
11. Check what the new website actually collects
The technology on the new website may not be the same as the old one.
Perhaps you’ve added analytics that weren’t previously there. There’s a new enquiry form. Marketing has introduced an email platform. Videos are being embedded from another service. A booking system, chat tool or advertising technology has been added.
Those changes can affect the information the website collects and the cookies or similar technologies it uses.
For a UK business, the redesign is therefore a sensible time to review whether the privacy information and cookie setup still reflect what the website actually does.
Don’t simply copy the privacy and cookie setup from the old site and assume it’s still appropriate.
The aim here isn’t to turn the redesign team into privacy lawyers. Where necessary, get appropriate legal or data-protection advice.
The website team’s responsibility is to make sure the question isn’t forgotten.
12. Test it like somebody who doesn’t already know the answers
The people building and approving a website have one major disadvantage when testing it.
They already know how everything works.
They know what the company does. They know where services are located. They know what a particular phrase means. They know what happens after somebody fills in a form.
A new visitor doesn’t.
Alongside the normal technical testing, give the website to somebody who wasn’t involved in creating it.
Ask them what they think the company does. Ask who they think it’s for. Give them a particular service to find. Ask what seems different about the business. Ask what they’d do if they wanted to make an enquiry.
You can learn a lot from watching somebody use the website without helping them.
Then do the technical checks.
Test phones and larger screens. Check relevant browsers. Submit every important form and make sure the emails arrive. Follow buttons and links. Test downloads. Check images. Review accessibility. Look at performance. Check cookie controls and any third-party integrations.
Don’t just test whether the website exists.
Test whether it works.
13. Launch with a plan for what happens if something goes wrong
A launch shouldn’t depend on optimism.
Before replacing an established website, have a current backup and know how it would be restored. Know who has access to the hosting, domain and website if something unexpected happens.
Once the new site is live, test it again in the live environment.
Submit the forms. Check analytics. Test important redirects. Verify that search engines aren’t being blocked by a development noindex instruction. Check canonical URLs and the XML sitemap. Look for broken links and unexpected 404 errors.
If the website already receives meaningful organic traffic, monitor Google Search Console and analytics after launch rather than assuming the migration is complete because the homepage appears.
Some search movement can happen while Google recrawls and reindexes a significantly changed website. The important thing is being able to distinguish normal processing from a genuine problem.
And don’t declare the launch complete at 9:02am because somebody typed the domain into their phone and the homepage appeared.
The first few days are part of the launch too.
14. Decide what you’re deliberately leaving behind
A redesign isn’t only about what goes onto the new website.
It’s also an opportunity to stop carrying things that no longer make sense.
You don’t automatically need the old navigation because customers are supposedly used to it. You don’t need every page simply because it already exists. You don’t need to keep service names the business no longer uses or copy written for customers you don’t particularly want anymore.
You don’t need the stock photographs everybody has disliked for six years.
And you don’t necessarily need every feature, integration or piece of technology that found its way onto the old site.
Be careful here. As we’ve already covered, existing content and URLs can have value that isn’t immediately obvious. Removing something should be a decision, not an accident.
But once you’ve understood that value, you’re allowed to leave things behind.
A redesign isn’t an obligation to recreate the old website more beautifully.
It’s an opportunity to decide what the website should be now.
The website redesign checklist°
If you’re preparing to redesign an established business website, this is the shorter version to work through before and during the project.
Understand°
- Can we clearly explain why we’re replacing the existing website?
- What has changed in the business since it was built?
- Which customers and work do we want more of?
- What do we want the new website to do better?
Protect°
- Which existing pages receive useful search traffic?
- Which URLs, content and external links have value?
- What currently generates enquiries or helps the sales process?
- Have we identified anything that shouldn’t disappear accidentally?
Decide°
- What are we keeping, improving, combining and removing?
- Does the new sitemap reflect today’s business rather than the old navigation?
- Do we know what each important page needs to communicate?
- What evidence will support the claims we’re making?
Prepare°
- Do we have the content, photography and brand assets we need?
- Do we control the domain, DNS, hosting and relevant accounts?
- Do we have analytics and Search Console access?
- Do we understand any third-party integrations?
- Is there a clear person responsible for approvals?
Test°
- Has the website been tested on relevant devices and browsers?
- Have forms, emails, downloads, links and integrations actually been tested?
- Has someone outside the project tried to use it?
- Have accessibility, performance, security and privacy/cookie requirements been reviewed?
Launch°
- Are changed URLs mapped to appropriate redirects?
- Are internal links, canonicals and the XML sitemap correct?
- Have development indexing restrictions been removed?
- Is analytics working?
- Is there a current backup and a way to restore it?
- Does somebody know what needs monitoring after launch?
You don’t need every answer on the first day.
But these are the questions that should be answered somewhere between deciding to replace the website and telling the world the new one is finished.
Because the success of a redesign isn’t simply whether the new website looks better than the old one.
It’s whether it represents the business better, protects what was already working and gives the company a more useful website for wherever it’s going next.