Website redesign or rebuild: how to tell which one your site actually needs

Once you've decided your website needs work, the next question is what shape that work should take. Most teams default to "we need a new website" without distinguishing between two very different projects: a redesign and a rebuild.

They cost different amounts, take different lengths of time, and solve different problems. Picking the wrong one is one of the more expensive mistakes a marketing team can make, because you usually only find out after the money has been spent.

What each one actually means

A redesign keeps the underlying technology and rebuilds the visual layer on top of it. The CMS stays, the hosting stays, the database stays, the basic site structure usually stays. What changes is how the site looks, how the pages are laid out, the typography, the imagery, the calls to action, and often the messaging.

A rebuild starts from scratch. New theme, new code, often new CMS, new hosting, new everything. The content might carry across, the URL structure might be preserved, but the foundation is replaced.

A redesign typically costs between £4k and £12k for a service business site. A rebuild typically costs between £12k and £35k. The time difference is usually six weeks versus three to four months. The disruption to the team is roughly proportional.

When a redesign is the right call

A redesign is the right answer when the foundation is sound but the surface isn't doing its job. The site loads quickly, the CMS works fine, the developer who built it left things in a reasonable state, and the back end isn't holding anyone back. The problem is specifically how the site looks, reads, or converts.

Common scenarios where a redesign fits cleanly: the design feels visually dated but the structure is fine, the messaging needs a refresh because the business has evolved, the homepage isn't converting and needs reworking, the brand has been updated and the site needs to catch up, or the mobile experience needs attention.

The defining feature of all of these is that you're not fighting the platform. You're just changing what sits on top of it.

When a rebuild is the right call

A rebuild is the right answer when the foundation itself is the problem. The site is slow in ways that can't be fixed without rewriting it. The CMS is outdated or unsupported. The codebase is held together with workarounds. Every small change is expensive because the original build was put together in a way that makes maintenance painful.

Other rebuild signals include: the site was built on a platform you can no longer get good support for, plugins or dependencies are no longer maintained, security has become a recurring problem, or the structure of the site is so far from how the business now works that no amount of redesign can fix it.

The honest test is whether you can imagine your developer saying "yes, we can do that" without flinching when you ask for changes. If small changes are reliably hard, the foundation has failed and a rebuild is justified.

The middle case nobody talks about

There's a third option that gets overlooked, and it's often the right answer for sites that don't fit cleanly into either bucket. A targeted rebuild of specific sections, with the rest of the site left alone.

This works when most of the site is fine but a few critical pages or templates are broken in ways that can't be patched. You rebuild those specific parts properly, leave the working parts where they are, and revisit the rest in a future phase. The cost sits between a redesign and a full rebuild, and the disruption is much lower than a full project.

The reason it gets overlooked is that it's harder to scope and it doesn't fit neatly into a sales conversation. Agencies tend to recommend either a redesign or a rebuild because those are the two options the industry has standardised on. The middle path is usually what the site actually needs.

How the wrong choice goes wrong

Picking a redesign when you needed a rebuild is the more common mistake. The new design looks great in mockups, the developer puts it on top of the existing foundation, and within six months the same problems start surfacing. Pages are still slow because the underlying code didn't change. Small updates are still painful for the same reasons they were before. The conversion improvements you saw at launch erode as the site drifts.

The team then has the same conversation about needing a new website 18 months later, having spent £8k on a redesign that didn't solve the actual problem. The eventual rebuild ends up costing more than it would have done in the first place, because you're now also undoing the redesign work that didn't survive.

Picking a rebuild when you only needed a redesign is the other failure mode, and it's usually pushed by agencies who'd rather sell a bigger project. You spend three months and £20k replacing a foundation that was actually fine, and the conversion gains you get could have come from a £6k redesign in six weeks.

Both mistakes are recoverable, but both are expensive. The whole point of asking the question properly upfront is to avoid them.

The diagnostic questions worth asking

A few questions usually clarify which side you're on.

When was the site originally built, and what was it built on? A four-year-old WordPress site on a current theme is probably a redesign candidate. A nine-year-old site on a long-abandoned platform is almost certainly a rebuild.

How easy is it for your developer to make small changes? If the answer is "not very", that's a foundation problem and a redesign won't fix it.

Are you changing CMS, hosting, or platform as part of the work? If yes, you're rebuilding by definition, even if you're calling it a redesign.

How fast is the site currently? If it's failing Core Web Vitals in ways that can't be solved by image optimisation or removing plugins, the underlying code is the problem. That's a rebuild conversation.

Has the structure of the business changed in a way that requires new content types, new sections, or a meaningfully different sitemap? If yes, you're closer to a rebuild than a redesign, because the changes go deeper than the visual layer.

What good agency advice looks like here

The right agency answer to this question is rarely "definitely redesign" or "definitely rebuild" without first looking at the site. Anyone giving you a confident recommendation in a sales call before they've actually opened the back end is guessing.

Good advice usually looks like a short technical audit, an honest conversation about what's working and what isn't, and a recommendation that reflects what the site actually needs rather than what the agency would prefer to sell. The recommendation might be a redesign, a rebuild, or the targeted middle path. It might also be "don't do anything yet, here are three things to fix on your current site for £2k that will buy you another year".

If you're in this conversation with an agency and they haven't asked to look under the bonnet before recommending, that's a warning sign worth paying attention to.

How to actually decide

Run the diagnostic questions. If the foundation is sound and the problem is the surface, you're looking at a redesign. If the foundation is the problem, you're looking at a rebuild. If the answer is unclear, the targeted middle path is usually worth exploring before committing to either extreme.

If you're not sure which side of the line you're on, our web design service page covers how we approach both types of project, and our recent posts on when to redesign your website and how much a website actually costs in 2026 cover the surrounding decisions.