The 30-second version
“We need a new website” is a conclusion, not a diagnosis. Some aging sites have good bones: the content earns traffic, the structure makes sense, and what’s actually wrong is speed, dated design, or a weak mobile layout. Those get fixed, not demolished. Other sites are genuinely done: the platform is a dead end, mobile is broken at the foundation, the structure fights SEO, or the thing has been patched so many times nobody dares touch it. The skill is telling the two apart before money moves, because the sunk-cost trap runs in both directions: owners who keep pouring effort into a doomed site, and owners who torch a site whose rankings took five years to earn.
Signals that say keep and fix
Good bones, bad speed. The site is organized sensibly and reads well; it’s just slow. A bloated theme, oversized images, a decade of accumulated plugins. Speed is fixable without demolition, and fixing it preserves everything else that works.
Content that earns, design that dates it. Check your analytics. If pages on your site rank and pull steady organic traffic, that content is an asset with years of accumulated equity behind it. A dated wrapper around working content calls for a redesign that leaves URLs, headings, and copy substantially intact.
Conversion problems, not foundation problems. Traffic arrives and doesn’t call. That’s usually layout, messaging, and missing trust signals: fixable on the existing foundation, often page by page.
It’s you, not the site. Sometimes the honest diagnosis is neglect. A structurally fine site with three-year-old photos and a stale services list needs an afternoon of updates, not a proposal.
Signals that say rebuild
Platform dead end. The site runs on an abandoned builder, a proprietary system only one long-gone developer understood, or a stack that no longer gets security updates. You can’t renovate a foundation the ground is swallowing.
Mobile is broken, not just imperfect. If the phone experience is a shrunken desktop site, or the layout genuinely fails on small screens and the theme can’t be taught otherwise, you’re rebuilding either way. The only question is whether you admit it now or after another year of lost visitors.
The structure fights SEO. URLs that make no sense, duplicate pages competing with each other, no logical hierarchy, headings used as decoration. When the architecture itself is the ranking problem, patching pages is repainting a house with a cracked frame. This is where SEO and web decisions stop being separate conversations.
Nobody dares touch it. The tell we hear most often: “we don’t change anything because last time the whole site broke.” A website too fragile to edit has already failed; you’re just still paying to host it.
The sunk-cost trap runs both directions
The famous version is clinging: “we’ve put so much into this site, we can’t scrap it.” Time already spent is gone either way; the only question is which path forward earns more from here.
The version nobody warns you about is torching. A business gets redesign fever and throws away pages that took years to earn their rankings. The new site launches, URLs change without redirects, content gets “freshened” into oblivion, and organic traffic falls off a cliff the following month. The old site’s design was worthless; its search equity was not, and both left in the same dumpster.
Both traps have the same cure: inventory what the current site has actually earned (rankings, traffic, links, pages that convert) before deciding anything. Whatever performs is an asset, no matter how ugly its container.
How we actually decide
Our web design engagements start with that inventory, and the recommendation follows the evidence. When we rebuild, the performing content moves over intact: same URLs where possible, redirects where not, rankings treated as cargo to protect rather than furniture to replace. When we redesign, we’re explicit about what stays untouched and why. The goal in both cases is identical: keep what the site has earned, replace what’s holding it back, and be honest about which is which.
A useful rule of thumb: design problems and speed problems usually mean fix. Platform problems and structure problems usually mean rebuild. If you have both kinds, rebuild, but migrate the equity deliberately. And if you can’t tell which kind you have, that itself is a finding: get the diagnosis before you accept anyone’s proposal, especially from someone whose business model is proposals.
Common questions
How do I know if my content is actually earning anything? Google Search Console: free, and connected to your site in an afternoon. If pages get impressions and clicks from search, they’re earning. If the whole site is invisible, there’s less equity to protect and the rebuild decision gets easier.
Will a redesign hurt my rankings? A careless one can, badly. The risk isn’t redesigning; it’s changing URLs without redirects, deleting pages that rank, and rewriting content that was working. Handled deliberately, a redesign that improves speed and mobile experience usually helps.
How old does a site have to be before rebuilding is automatic? Age alone decides nothing. We’ve seen decade-old sites worth keeping and three-year-old sites beyond saving. The platform and the structure decide, not the birthday.
Can I do this in stages? Often, and it’s underrated: fix speed first, then the highest-traffic templates, then the rest. Staging only fails when the foundation itself is the problem, which is exactly what this diagnostic exists to catch.