Should you rebuild your website or fix it?

The quote is the visible cost and usually the smaller one. A rebuild also spends organic performance, because URLs change and rankings take months to recover even when redirects are done properly — and they are frequently not. It spends the conversion rate you already had, because a new design is a new set of assumptions and some of them will be worse. And it spends attention: six months during which the marketing conversation is about the website rather than about customers.

None of that argues against rebuilding when it is genuinely warranted. It argues for being clear about which category you are in, because the same money spent on three targeted fixes frequently produces more than a rebuild does, and without the recovery period.

Four situations justify it, and they share a characteristic: the problem is structural rather than cosmetic, so no amount of fixing reaches it.

• The platform cannot do what the business now needs, and extending it costs more than replacing it. A brochure site being asked to take bookings and payments is the common case.
• The information architecture fights the business. If the services you sell today cannot be found because the structure reflects what you sold five years ago, page-level fixes cannot resolve it.
• Accumulated technical debt makes changes risky. When nobody will touch the site because something broke last time, the maintenance cost has already exceeded the rebuild cost — it is just being paid in caution rather than in invoices.
• It cannot be made accessible or fast. Some builds are far enough from WCAG or from acceptable Core Web Vitals that remediation costs more than starting again, though this is claimed more often than it is true.

These are the reasons most often given, and each one has a cheaper answer that solves the actual complaint.

Before quoting a rebuild, we would normally propose these and see what remains. In a good proportion of cases what remains does not justify the larger project.

1. 1.Fix the pages that actually matter. Most sites have five pages carrying almost all the commercial weight. Rewriting those properly — proposition, evidence, price position, next step — moves more than redesigning forty pages nobody visits.
2. 2.Fix speed and mobile behaviour. Images, scripts and hosting account for the overwhelming majority of slow sites, and none of those requires rebuilding anything. Test on a mid-range Android on a real connection, not on a desktop over office broadband.
3. 3.Fix the structure of the navigation without touching the code. Many "we cannot find anything" problems are menu problems, and reorganising what is already there is a day of work rather than a project.

Assuming the decision is made properly, these are the failures that make rebuilds cost more than they should. All are avoidable and all are commonly missed.

• Redirects treated as a launch task. Every old URL needs mapping to its closest equivalent, and that work belongs in the design phase because it constrains the new structure.
• Content written last. It is the longest lead item on every web project and it is invariably scheduled as though it were the shortest.
• The conversion rate not being measured before launch. Without a baseline you cannot tell whether the new site is better, and everybody will have an opinion instead.
• Losing the pages that were ranking. Somebody should list the top fifty organic landing pages before anything is designed, and every one of them should survive in some form.
• Tracking rebuilt from scratch and quietly broken. Verify conversions with a real submission on launch day, not from the analytics dashboard a fortnight later.
• Accessibility deferred to a later phase. It is cheap when built in and expensive when retrofitted, and the later phase rarely happens.

There is a third answer that almost never gets proposed, and it is right more often than either of the other two: leave the site alone and spend the money on getting people to it.

A site converting adequately with very little traffic does not have a website problem. Rebuilding it produces a better-looking page that the same small number of people see, and the enquiry count barely moves. The same budget spent on search visibility, paid media or content would have changed the number, and the honest recommendation in that situation is to say so rather than to quote for the build.

The test is simple. Look at your monthly visitors and your conversion rate. If the conversion rate is reasonable for your sector and the visitor count is small, the constraint is upstream. Fix the site when the site is what is failing, and not because it is the thing that is easiest to point at.