The short version
- Looking dated is the most common reason given for a rebuild and one of the weakest.
- Rebuild when the foundations are the problem: speed you cannot fix, a CMS your team avoids, or a structure that stops the site ranking.
- Redesign when the foundations are sound and the site simply says the wrong things.
- Sometimes the honest answer is neither, and the problem is that nobody is driving traffic to it.
Websites rarely get replaced because someone measured something. They get replaced because a director looked at it on their phone and did not like it, or because a competitor launched something smarter, or because it has simply been a few years.
Those are understandable prompts and poor briefs. Here is a more useful way to decide, and what each answer actually involves.
Signs you need a rebuild
A rebuild replaces the foundations: the platform, the theme or codebase, and usually the structure. It is the expensive option and it is the right one when the problem is structural.
It is slow and cannot be made fast
If the site takes several seconds to become usable on a phone and the cause is a heavy theme stacked with plugins, that is not tunable. Speed affects rankings and affects conversion considerably more. A site that cannot be made fast is a site that will keep costing you.
Your team avoids editing it
A telling symptom. If updating a page means emailing an agency, or if your team have quietly stopped adding content because the editor fights them, the site has stopped being an asset and become an obstacle. That is a foundations problem.
It cannot connect to anything
Enquiries arriving as emails that somebody rekeys into a CRM. Stock or availability maintained in two places. No way to get the source of an enquiry into the system where the sale is recorded. If the site cannot talk to the systems the business runs on, patching it does not help.
The structure is why it does not rank
Everything crammed into four pages, one of which is called Services. Nothing for the individual things people actually search. No sensible place to put new content. You cannot fix that by redesigning the pages; there need to be different pages.
Nobody can tell you what it produces
No conversion tracking, no call tracking, no source on any enquiry. Strictly this is fixable without a rebuild, but in practice it tends to arrive alongside everything else on this list.
Ask whoever maintains the site to add a new service page while you watch. If it takes ten minutes, the foundations are fine. If it needs a developer, or it cannot be done without breaking the layout, you have your answer.
Signs a redesign is enough
A redesign keeps the platform and the structure and changes how the site looks and what it says. It is faster, cheaper, and the right answer more often than agencies tend to admit.
- The site works, it just does not persuade. Loads quickly, easy to edit, connects to what it needs to. It simply does not explain the proposition well or ask for the enquiry clearly.
- The design is genuinely dated but nothing else is. Fine. That is a real reason, it is just a smaller job than a rebuild.
- The content is wrong rather than the container. The business has moved on and the site describes what it did five years ago.
- Two or three pages do all the work. If the homepage, one service page and contact carry the site, rewriting those three beats replacing forty.
Signs you need neither
This is the one nobody gets told, and it is worth saying plainly.
Nobody is driving traffic to it. If the site gets fifty visitors a month, the problem is not the site. Rebuilding it will produce a better site that fifty people see. The money belongs in search or advertising first.
Enquiries arrive and are not followed up. We have watched businesses commission a rebuild while sitting on enquiries nobody called back for three days. Fixing response time is free and works faster than any rebuild.
The enquiries are the wrong sort. That is usually a targeting and qualification problem, and both are cheaper to fix on the existing site.
It was rebuilt eighteen months ago. If it is still not working, something other than the website is wrong, and rebuilding again will not find it.
Not sure which of the three you are? Send us the URL. We will tell you what the current site earns, what is actually holding it back, and which of redesign, rebuild or neither we would recommend. Including when the answer is that you should spend the money elsewhere.
What each one actually involves
| Redesign | Rebuild | |
|---|---|---|
| Platform | Kept | Replaced or rebuilt |
| URLs | Mostly unchanged | Often change, so need mapping |
| Ranking risk | Low | Real, and manageable with planning |
| Content | Rewritten | Rewritten and restructured |
| Integrations | Usually untouched | The part that moves the cost most |
| Typical driver | It does not persuade | It cannot do what we need |
The question to ask before committing
Whatever you are told, ask this: what will be measurably different twelve months after launch, and which number will show it?
If nobody can answer, the project has no definition of success and will be judged on whether people like the look of it. That is how businesses end up rebuilding again in three years.
Common questions
How often should a website be replaced?
There is no schedule, and treating it as one is how money gets wasted. A well-built, maintained site serves for years with periodic refreshes. What ends a site’s life is the business outgrowing what it can do, not its age.
Can we rebuild in stages?
Occasionally, for very large sites. For most businesses it creates two structures running at once, which is harder to redirect and harder to diagnose. All at once is usually cleaner.
Our site is on a builder platform. Does that force a rebuild?
Only if the platform is genuinely the constraint. If it is fast, editable and connects to what you need, changing it is expensive disruption. The limits usually appear around real integration and serious search performance.
What does a redesign cost compared with a rebuild?
A redesign is typically a fraction of a rebuild, because the expensive parts are integration, migration and structure, and a redesign touches none of them. We set out what drives the numbers in our guide to what a website should actually cost.
