Why most website redesigns fail

A client came to us last year after their redesign went sideways. They’d spent four months and a significant budget rebuilding their site from scratch. New look, new structure, new CMS. It launched on a Friday, and by Monday their organic traffic had dropped 40%.

Nobody had set up redirects. Hundreds of indexed URLs were returning 404s. The structured data that had taken two years to build up in Google’s index was gone overnight.

This isn’t unusual. Most redesigns fail, not because the new design is bad, but because the process ignores things that matter more than aesthetics.

Mistake 1: ignoring what’s already working

The most common redesign impulse is “the site looks dated, let’s start fresh.” But “start fresh” often means throwing away things that were quietly doing their job.

Before touching anything, you need to know:

• Which pages drive the most organic traffic?
• Which pages convert visitors into leads or customers?
• What search queries bring people to the site?
• Which content gets shared or linked to?

Google Analytics and Search Console will give you this data. If you redesign without looking at it, you’re guessing. And guessing with a business website is expensive.

Here’s a real scenario. A services page ranked #2 for a valuable local keyword. During the redesign, the content was rewritten to be “more modern,” the URL was changed, and no redirect was added. That single page was generating roughly €3,000/month in leads. It took six months to recover the ranking, if it even fully recovered. That’s potentially €18,000 in lost business from one oversight.

Mistake 2: the redirect graveyard

When you change URLs during a redesign and don’t set up 301 redirects, every link pointing to the old URLs becomes a dead end. This includes:

• Google’s index (your search rankings)
• Links from other websites (your backlink profile)
• Social media posts and shares
• Printed materials, business cards, email signatures
• Bookmarks your customers saved

A proper redirect map should be one of the first deliverables in any redesign project:

Every single URL from the old site needs a destination. No exceptions. For large sites, this is tedious work, but skipping it is one of the most expensive shortcuts you can take.

In Hugo or any static site generator, redirects are straightforward to implement:

# In hugo.toml or a redirect config
[[redirects]]
  from = "/services/web-design"
  to = "/services/websites"
  status = 301

[[redirects]]
  from = "/blog/2024/seo-tips"
  to = "/insights/seo-tips"
  status = 301

Mistake 3: redesigning based on opinions instead of data

“I think the homepage needs more white space.” “My wife doesn’t like the blue.” “Our competitor has a video background.”

These are opinions, not strategy. A redesign should be driven by:

• User behavior data: Where do people click? Where do they drop off? What do heatmaps show?
• Conversion data: Which pages convert and which don’t? What’s the path to conversion?
• Search data: What content brings organic traffic? Which keywords are you close to ranking for?
• User feedback: What do actual customers say about using the site? Have you asked them?
• Accessibility audits: What’s failing WCAG compliance? What does screen reader testing reveal?

I’ve seen redesigns where the entire navigation was restructured because the CEO didn’t like it, ignoring the fact that user testing showed people found things easily. Personal preference is not user research.

Mistake 4: scope creep that never ends

A redesign that starts as “update the visual design” almost always evolves into “also let’s add a blog, rebuild the contact form, integrate a CRM, add a client portal, and maybe an e-commerce section.”

Each addition doubles the timeline and introduces new complexity. Suddenly a two-month project is six months in and still not done.

The fix is disciplined phasing:

Phase 1 (launch): Visual refresh, redirect map, content migration, performance optimization. Get this live.

Phase 2 (month 2-3): New features like a blog section, improved contact flow, analytics refinements.

Phase 3 (month 4+): Larger additions like e-commerce, portals, integrations.

Ship the core redesign first. Then iterate. A perfect site that never launches helps nobody.

Mistake 5: not testing before launch

I’m consistently surprised by how many redesigns go live without proper testing. “It looks fine on my laptop” is not a testing strategy.

A pre-launch checklist should include:

• Cross-browser testing: Chrome, Firefox, Safari, Edge. At minimum.
• Mobile testing: Real devices, not just browser dev tools. Test on both iOS and Android.
• Form testing: Every form, every field, every validation rule, every confirmation email.
• Performance testing: Run Lighthouse, check Core Web Vitals, test on slow connections.
• SEO verification: Structured data validates, sitemap is correct, robots.txt isn’t blocking important pages, redirects work.
• Accessibility testing: Keyboard navigation, screen reader testing, color contrast, focus indicators.
• Content review: No lorem ipsum, no broken images, no placeholder text, all links work.

# Quick pre-launch checks you can automate
npx lighthouse https://staging.yourdomain.com --output=html
npx pa11y https://staging.yourdomain.com
curl -s -o /dev/null -w "%{http_code}" https://staging.yourdomain.com/old-url

If any of these checks fail, you’re not ready to launch.

Mistake 6: treating content as an afterthought

The new design is approved, the templates are built, and then someone asks “so who’s writing the content?” This happens far more often than it should.

Content should drive the design, not the other way around. The questions to answer before any visual design work begins:

• What pages does the site need?
• What’s the messaging hierarchy on each page?
• What calls to action matter most?
• What content already exists that’s worth keeping?
• Who’s responsible for writing new content, and what’s the timeline?

Designing layouts before you know what goes in them leads to generic templates that don’t serve the actual content. Or worse, it leads to content being crammed into boxes that weren’t designed for it.

How to do it right

A redesign that works follows a predictable process:

1. Audit first: Analyze current analytics, search performance, user behavior, and technical health. Know what you’re working with.
2. Set measurable goals: “Increase organic traffic by 20%” is a goal. “Make it look more modern” is a vibe.
3. Map every redirect: Before building anything, document where every old URL goes.
4. Content-first design: Write or outline the content, then design around it.
5. Phase the rollout: Launch the core redesign, then add features in iterations.
6. Test thoroughly: Every browser, every device, every form, every redirect.
7. Monitor aggressively post-launch: Watch analytics daily for the first two weeks. Compare traffic, rankings, and conversions to the pre-redesign baseline.

The goal isn’t a new website. The goal is a better-performing website. Those are very different things, and the process you follow determines which one you get.

Is it even time for a redesign?

Before committing to a full redesign, consider whether you actually need one. Sometimes the fix is targeted improvements rather than starting over:

• Slow site? Optimize performance without touching the design.
• Poor rankings? Improve content and technical SEO.
• Low conversions? Test different calls to action and page layouts.
• Outdated look? A visual refresh of colors, typography, and images can modernize a site without rebuilding it.

A redesign is a big investment. Make sure the problem actually requires one.

If you’re considering a redesign and want to talk through whether it’s the right move, we’re happy to chat . You can also see examples of our work or use our calculator to get a sense of what a project like this typically involves.