Website rebuild or refresh? How to tell which one you actually need

Everyone agrees the website is not good enough. Nobody agrees on why. One person wants it to look modern, someone else says it is slow, and the quotes you have collected range from a few thousand pounds for a facelift to a number that would fund a small hire. The question underneath all of it is the same one: rebuild or refresh? Working that out is worth an afternoon of your own time, because the answer decides everything else.

Website problems come in four kinds, and only two of them are fixed by design work. A site can look dated, be slow or broken, be structurally unable to do what the business now needs, or be perfectly fine except that it does not say anything convincing. Each has a different fix at a different price, and the most expensive mistake in this whole area is spending rebuild money on a problem that was never structural. You can tell which you have got in an afternoon, without hiring anyone.

The four kinds of "our website is bad"

The afternoon diagnostic

Seven tests, in order, on your own site. Do them properly rather than imagining the answers, because the point is to be surprised by at least one of them.

1. Open your site on your own phone, on mobile data, with the office wifi turned off. Count the seconds until you can read something, and the seconds until you can tap something without it jumping. If either is uncomfortable to sit through, you have a performance problem, and it is the one costing you the most, because most of your visitors are on a phone and they do not wait.
2. Do the main thing a customer does. Book, buy, enquire, find the opening hours, whatever it is. Time it and count the taps. Then hand the phone to someone who has never used it and watch them without helping, which is the hardest and most useful part of this entire exercise.
3. Ask someone on your own team to publish a new page. Not a text change, a genuinely new page. Sit next to them and do not help. If it needs a developer, a ticket, or an email to an agency, that is a structural finding, and it is the reason your site will be out of date again in a year regardless of what you spend now.
4. Read your homepage out loud to someone who does not know your business, then ask them two questions: what do we sell, and why would you choose us over the other one? If they cannot answer both, no amount of design will fix that, and you have found the cheapest and most valuable problem on this list.
5. Shrink the browser window on your computer until it is phone width, and go through the menu, a form and any table. Things that collapse, overlap or become untappable are a performance and build-quality finding.
6. Look at whether anyone is arriving at all. Open your analytics or search console. If the traffic is close to nothing, then design is not your problem yet, and a rebuild would be an expensive way to keep it that way.
7. Ask when the site last changed. If the answer is a year or more, and the reason is that changing it is a hassle, add that to the structural column, because it will happen again whatever the new one looks like.

Write down which tests failed. The pattern, not the individual failures, is the answer.

Signs it is a refresh

• The structure works and your team can publish without help. Tests three and seven passed.
• The complaints are about how it looks: dated typography, tired stock photography, a logo you have since replaced, colours that no longer match anything else you produce.
• It is reasonably quick, and the pages a customer needs are all there.
• You are about to rebrand, or you already have, and the site is now the last thing carrying the old identity. A design refresh is exactly this job.
• Nothing about the business has changed in a way the site cannot express.

A refresh is dramatically cheaper, faster, and lower-risk than a rebuild, and for a site whose bones are sound it is usually the right call. It also has a hard limit, and it is worth being blunt about it: a refresh makes a wrong model prettier. If the thing underneath cannot do what you need, new colours will not change that, and you will be having this conversation again in eighteen months with less patience and less budget.

Signs it is a rebuild

• Your own team cannot change it. The single most reliable indicator. A site only your supplier can edit decays by design.
• It is slow for structural reasons. Heavy page-building plugins, an overloaded template, images nobody can resize without a developer. If the slowness lives in how the site is built rather than in a handful of oversized photographs, no amount of tuning saves it.
• The business has outgrown the model. You had five services and now you have thirty, or you need accounts, bookings, quoting or payments where before you needed a phone number. The concepts the site was built around no longer match the business.
• Every change breaks something else. The clearest sign that the cost of each change is rising rather than falling, and that curve does not correct itself.
• It cannot be made accessible or secure as it stands. Structural accessibility problems and an unmaintained platform with known vulnerabilities are foundations, not decoration.
• Nobody can safely touch it. No staging copy, no way to preview a change, no version history, plugins nobody dares update.

Two or more of those, and you are almost certainly looking at a rebuild. One of them on its own is worth a second opinion before you commit to the larger number, and that is precisely what our redesigns and website rescues assessment is for.

Signs it is neither

This is the finding suppliers are least likely to volunteer, because it is the one that does not sell a project.

• The site loads quickly, works on a phone, and your team can edit it. Every technical test passed.
• People arrive, read, and leave without contacting you.
• Your homepage explains what you do but never says why anyone should pick you, what it costs, or what happens next.
• There is no obvious next step on any page, or there are eleven.
• You cannot say, in one sentence, who the site is for.

If that is your pattern, the problem is the offer and the words, and it is fixable for a fraction of either project. Rewrite the homepage around what a customer is actually deciding. Put prices, or at least ranges, somewhere on the site. Say who you are not for. Then look again in a month, because a rebuild that starts from unclear positioning simply produces a better-looking version of the same confusion.

What each one costs, roughly

A refresh is a design and front-end job on an existing structure, so it sits at the bottom of any build budget and often finishes in weeks rather than months. A rebuild is a build, priced like one, and the range depends on the same five things that drive any website price: how many unique page designs, what it connects to, who writes the words, what your team needs to be able to change, and who maintains it afterwards. Our guide to what a website costs and what moves the price sets out those bands in detail, along with what usually gets left out of a quote.

The content and offer fix costs least of all, and is the one most often skipped, because it involves deciding things about your business rather than approving designs.

Not for you if

The next step

Run the seven tests and write down what failed. If the failures are cosmetic, get a refresh quoted. If tests one, three and seven failed together, get a rebuild quoted and ask specifically what your team will be able to change afterwards. If everything technical passed and people still leave, put the budget into your words and your offer before you put it into pixels.

And whichever it is, ask any supplier to tell you which of the four categories they think you are in, and why. A supplier who reaches the same conclusion you did from the same evidence is one you can trust with the larger number.