Every year or two, someone in a leadership meeting says the website looks dated. Heads nod. A budget gets floated. Six months later there's a new site, a new set of colors, and roughly the same pipeline.
"Dated" is a description, not a diagnosis. A rebuild is a large, slow, high-risk project that fixes a specific set of problems — and leaves most others untouched. Before you approve one, it's worth spending a week finding out what's actually broken. Here are five tests we run before recommending anyone spend money on a redesign.
Test 1: Is the Failure in Traffic, Conversion, or Follow-Up?
Think of your site as three stages: people arriving, people acting, and what happens after they act. Each stage has a different fix, and only one of them is really a website project.
If qualified people aren't arriving, the constraint is content, search visibility, and how well your pages answer real buying questions — including for AI answer engines that now summarize before anyone clicks. A prettier site doesn't move that needle.
If people arrive and leave without acting, that's genuinely a design and messaging problem, and a rebuild may help.
If people submit forms and nothing good happens next, the problem is downstream: routing, ownership, speed of response, CRM hygiene. We've seen more revenue lost in that third stage than in the first two combined, and a redesign fixes none of it.
Test 2: Can Your Team Change the Site Without a Developer?
Ask your marketing lead how long it takes to publish a new page. If the answer involves a ticket, a queue, or a phone call to an agency, you have a publishing problem, not a design problem.
Sites decay when editing is expensive. Content ages, offers go stale, product pages fall out of date — not because anyone chose that, but because changing anything costs a favor. A rebuild only fixes this if the new build explicitly makes editing a first-class requirement. Many rebuilds make it worse by delivering a beautiful, brittle site that only the original developer understands.
Test 3: Is the Slowness Structural or Situational?
Speed is a business problem, not a technical vanity metric. Slow pages lose mobile visitors, drag down search performance, and make good marketing work harder than it should.
But slow has two causes. Situational slowness is uncompressed hero images, a dozen tracking scripts, fonts loading late, no caching. That's often a few days of focused work. Structural slowness is baked in: a page builder that ships enormous markup on every page, a theme that loads every feature whether you use it or not, an architecture that can't be optimized without replacing it.
Run a real page-speed audit on your top five landing pages before deciding. If the fixes are situational, do them now and revisit the rebuild question in a quarter with better data.
Test 4: Do You Have a System or a Pile of One-Off Pages?
Count your page templates. Then count your pages. If almost every page is bespoke, your cost to launch anything new is permanently high, and consistency will keep drifting.
This is one of the strongest genuine arguments for a rebuild: replacing a pile of one-offs with a small set of reusable, well-built templates. The payoff isn't aesthetic — it's that the next fifty pages take hours instead of weeks.
Test 5: Where Does a Form Submission Actually Go?
Submit a form on your own site and follow it. Does it land in a shared inbox someone checks twice a day? Does it create a CRM record with the source attached? Does a rep get notified? Can you tell, later, which campaign produced the deal?
This is integration work, not web work, and it's usually the cheapest high-return fix available. The site is the front door; the value comes from what the building does with the visitor. If your marketing spend can't be traced to closed revenue, no amount of new design will make that spend defensible.
When a Rebuild Is Genuinely the Right Call
Sometimes the answer really is start over. The clearest signals:
- The platform can't do what the business now needs — commerce, multiple languages, gated content, customer portals.
- Core dependencies are unsupported or unpatched, making it a security liability.
- The content model is wrong, so new page types require workarounds every time.
- Nobody available to you can safely maintain the technology it's built on.
Those are structural. Everything else deserves scrutiny first.
Rebuild Without Resetting Everything
If you do rebuild, protect what already works. Keep your URL structure or map redirects deliberately. Migrate the content that earns traffic instead of rewriting everything from scratch. Take baseline measurements of speed, conversion, and search performance before launch so you can prove the new site is better rather than assuming it. And roll out template by template where you can, starting with the pages that carry the most traffic.
The goal isn't a new website. It's a lead-generation system that your team can operate, that hands clean data to your CRM and ERP, and that gets faster and better every quarter instead of aging quietly toward the next redesign.
If you're weighing a rebuild and want an honest read on whether it's the right investment, we're happy to walk your site and your funnel with you and tell you what we'd actually do first. Reach out — that conversation is usually worth more than the audit fee anyone would charge for it.