A website redesign is a money decision, not a taste decision
Before anything else, write down what the new site should do that the current one does not. More enquiries. Bookings without phone calls. Working properly on a phone. Being findable for a service you now sell. If the honest list is that it should look newer, you are about to spend a project budget on a feeling.
That is not a reason to keep an old site forever. Looking dated does cost trust, particularly for anyone choosing between you and a competitor on a Sunday evening. But looking dated is a symptom, and it is worth knowing which disease you have before you pay for surgery.
The rest of this article sorts the reasons that justify a rebuild from the ones that are cheaper to fix, and covers the part everybody forgets until the traffic drops.
Signs that it is time
Six signs, any two of which usually justify a rebuild.
- It does not work properly on a phone. Not merely small, but broken. Sideways scrolling, buttons that miss, text you have to pinch.
- The business has moved on. The site describes services you dropped and says nothing about the ones that now pay the wages.
- Nobody can update it. The platform is abandoned, the person who built it has gone, or every change needs a specialist.
- It is slow and nothing fixes it. Especially when the slowness comes from the platform and its plugins rather than from one heavy image.
- It cannot do what you now need. Bookings, a second language, a shop, a members area, and the current structure cannot hold them.
- It is a security liability. Unpatched software, no updates available, warnings you have learned to ignore.
Notice that five of the six are about function. Appearance is real, but it is rarely the reason on its own.
Signs that you only need a fix
Plenty of sites that feel tired are structurally sound and need a week of work rather than a project. The tell is that your complaints are about specific pages rather than about the whole thing.
Fixable without rebuilding. Outdated text and prices. Bad or missing photographs. A contact form that sends to an address nobody reads. Slow loading caused by huge images. Missing opening hours. Titles and descriptions that make your pages look like nothing in search results. A confusing menu with fourteen items.
A dental clinic in Jyvaskyla whose site looks ordinary but loads fast, works on a phone and ranks for its town does not need a redesign. It needs three better photographs, honest prices and a booking button that is visible without scrolling. That is a fraction of the cost and often a bigger jump in enquiries.
The test that settles most arguments
Open your site on your own phone, on mobile data, standing outside. Then try to do the thing a customer would want to do. Find a price. Book. Call. Find out whether you serve their town.
Count the taps and the seconds. If it takes more than about five seconds to load and more than three taps to reach the answer, you have found the problem, and it is usually not the colour scheme.
What a redesign can cost you in Google
This is where redesigns go wrong quietly. A site that has been live for years has accumulated addresses that Google knows, links from other sites, and a track record for certain searches. A rebuild that changes every address and throws away the old text can lose a large share of that overnight, and the drop shows up two weeks after launch when everyone has stopped watching.
The damage comes from a short list of causes. Page addresses that changed with no redirect from the old one to the new. Text that was shortened because it looked long, taking with it the words people searched for. Pages quietly dropped because nobody could see why they existed. A staging site left open to search engines, or a live site left closed to them after launch.
None of that is hard to prevent. All of it is hard to recover from, because the recovery depends on other people's sites and on time. If search matters to your business, our starter guide to search for a small business is worth reading before you brief a redesign rather than after.
The address map is the single most important document
Before a line of design work starts, list every address on the current site and decide, for each one, whether it stays the same, moves to a new address, or disappears. Every move needs a permanent redirect from the old address to the new one. Every disappearance needs a decision about where that visitor should land instead.
Pull the list from your analytics and your search tools, not from memory. Almost every site has pages the owner has forgotten that still bring visitors, often an old article or a page about a service you barely mention any more.
Check the redirects on the day of launch and again a week later. A redirect that was configured but never tested is the most common reason for a traffic drop that nobody can explain.
How to keep what already works
Find out what is working before you replace it. Your analytics will show which pages bring visitors and which ones lead to contact. Your search console will show which searches bring people in. Your inbox will show which questions keep arriving.
Those pages and those words are assets. Keep the addresses, keep the substance of the text, and improve around them. A page that brings you twenty enquiries a year is not a candidate for shortening because it looks wordy in the new layout.
Keep the practical things too. The phone number formatted so a thumb can dial it. The map that works. The form that reaches the right person. Redesigns lose these constantly, because they are invisible in a design review and obvious to a customer.
Measure before you touch anything
Write down today's numbers before the project starts, because afterwards you will have no baseline and every argument becomes an opinion. Visitors per month, enquiries per month, phone calls if you can count them, and the ten searches that bring the most people.
Take screenshots of the current site too. In six months nobody will remember what it looked like, and the comparison is useful both for judging the work and for explaining the spend to whoever approved it.
Redo the content before the design
The most common redesign failure is a beautiful new site carrying the same tired text. The layout changed and the message did not, so the results do not change either.
Start with the message. What do you actually do, for whom, where, and why should someone choose you over the firm in the next town? Answer those in plain language, then let the design carry the answer. Design is very good at making a clear message stronger and hopeless at rescuing a vague one.
This also protects the search side, because rewriting text with the words your customers use is exactly what improves visibility. Cutting text to fit a layout is what harms it.
When the platform is the real problem
Sometimes the site is fine and the thing underneath it is the problem. The symptoms are recognisable. Updates that break the layout, a pile of plugins each maintained by a different stranger, a monthly bill that grows, and a fear of touching anything.
In that case the decision is not about design at all. It is about moving to something you can maintain and afford, and the design can largely come along. Many rebuilds are really migrations wearing a new coat.
Ask any supplier what happens to your running costs and your dependencies after the move. A rebuild that swaps one fragile stack for another has solved nothing and cost a project budget doing it.
Doing it in stages instead of one big launch
A full rebuild is not the only shape available. For a site that mostly works, staged improvement often returns more per euro and carries far less risk.
A sensible order. Fix speed and the phone layout first, because they affect everything. Then rewrite the pages that bring the most visitors. Then the new photographs. Then the visual refresh, once you know which pages matter. Then the new functionality, when the need is proven.
What to check in the first month after launch
Put four checks in your calendar rather than trusting that it went well. In the first week, test every form and every redirect yourself, on a phone and on a computer. Confirm the site is visible to search engines and that the staging version is not.
In the second and fourth week, compare visitors and enquiries with your baseline. A small dip in the first fortnight is normal while search engines recrawl. A steady decline after a month is a problem, and the cause is nearly always redirects or removed text.
Also read what the site now says about your prices and your opening hours, slowly, out loud. Launch day errors in those two are the ones customers notice first. Keeping the site accurate afterwards is ordinary website upkeep, and it is the difference between a redesign that ages well and one that is tired again in two years.
When not to redesign at all
Do not redesign because three years have passed. There is no expiry date on a site that loads fast, works on a phone, says true things and brings you work. Replacing it resets a track record you spent years building.
Do not redesign to fix a business problem that lives elsewhere. If enquiries dropped because a competitor opened nearby or because your prices moved, a new website will not answer that, and the money would do more in the part of the business that is actually struggling.
And do not start a redesign without knowing what the current site does well. If you would like an outside read on that before you commit, our free review of your website will tell you plainly which of your complaints need a rebuild and which need an afternoon.