I’ve been handed websites with thousands of pages—some dating back over a decade, others duplicated so many times that no one on the internal team could explain which version was current. Navigation menus with 15 dropdowns, entire departments with self-built microsites, and content indexed in ways that made sense only to the teams that created them.
This is where many large redesign projects begin—not with a clean slate, but with layers of legacy architecture, each carrying the imprint of decisions made under different priorities, leadership, or urgency.
These aren’t just outdated websites. They’re digital representations of organizational sprawl. And transforming them isn’t a design challenge—it’s a systems-level project that touches people, processes, and platforms.
You don’t redesign legacy sites—you untangle them
The older and larger a website becomes, the more its architecture reflects internal structures rather than user logic. Departments carve out their own styles. Old campaign pages linger. Navigation grows, but not in ways that help people find what they need. The result? A patchwork system that seems to grow but never evolves.
Legacy sites aren’t just collections of outdated design—they’re symptoms of outdated thinking. The site still works, but not in a way that supports the user, the brand, or even the teams maintaining it.
The goal of a redesign isn’t simply a new visual layer. The goal is to reconnect the structure to purpose—to shift from internal logic to user logic, from departmental silos to unified pathways.
Here are five rules for a successful website redesign.
1. See the whole system before changing any one part
Jumping into redesign before deeply understanding the existing architecture is a common mistake. What looks like poor UX may actually be the result of a fragile backend dependency. What seems like redundant content may serve different audiences—or was created because the original was too hard to find.
Early in the process, build detailed maps and audit every layer of the site by asking:
- How deep is the content tree?
- Where do visitors actually land?
- How is content reused across departments or regions?
- What’s broken but still relied upon internally?
Without this foundation, every decision downstream becomes reactive.
2. Legacy content requires strategic curation, not bulk migration
I’ve heard it many times, “We were just looking to migrate everything.” One of the biggest risks in large-site redesigns is the temptation to migrate everything “as is.” But carrying content forward without purpose just creates clutter no one understands.
Instead of defaulting to full migration, we usually recommend a triage process:
- Review mission-critical pages that drive core business goals or have high user engagement.
- Rewrite and consolidate valuable but outdated or inconsistent web pages.
- Retire pages that are redundant, orphaned, or no longer aligned with strategy. Create 301 redirects to new and relevant pages.
This isn’t just cleanup—it’s a realignment of messaging, voice, and relevance.
3. Design systems work when matched to the content reality and user demand
A scalable design system is essential—not just visually, but structurally. It needs to accommodate the volume, variety, and velocity of content flowing through the site.
If the site has hundreds of recurring content types (reports, bios, specs, case studies), then repeatable components with built-in flexibility become the engine of consistency. Typography, icons, colors, and grids must map to how content lives—not just how it looks.
I’ve seen teams over-design pages with no consideration for how future content will be authored, leading to either visual breakage or over-reliance on designers for everyday updates. That’s not scalable.
4. Navigation must earn its place
Navigation is often treated as a checklist—list all the departments, link all the services. But on a large-scale site, that thinking leads to overwhelming menus and cognitive overload.
The better process favors subtraction over addition. Build navigation from user pathways, not organizational charts. Then test it by asking:
- Can a first-time visitor find what they need without using search?
- Do the menu labels reflect user vocabulary, not internal jargon?
- Is the mobile experience navigable without frustration?
I’ve learned to question everything. Just because a link has existed for 10 years doesn’t mean it deserves space in the new system.
5. Prototyping at scale is the only way to see the cracks
One of the turning points in a recent project was catching a user interaction issue that broke only on a specific mobile device, under a very specific screen width—caused by the interplay between a sticky menu, a modal, and an animation script. On paper, it looked clean. In the prototype, it failed.
That’s why I push for prototyping real data early—full content sets, edge cases, uncommon breakpoints. You don’t need to build the full site—but enough to see what might break at scale.
How to apply this to any organization facing a redesign
If you’re overseeing a website with years of content, multiple internal stakeholders, or inherited infrastructure, here’s what to take away:
- Begin with diagnostics, not design. You need to know what you’re fixing before proposing how.
- Design for editors and content owners as much as users. A system that’s hard to maintain won’t scale, no matter how elegant it looks.
- Redesign the logic, not just the layout. When navigation, language, and pathways match user intent, the experience improves even before visual polish is added.
- Test for how people actually use it. Real users reveal blind spots internal teams will always miss.
Closing reflection
Redesigning a large legacy website is often one of the most revealing exercises a company can go through. It exposes inefficiencies, misalignments, and outdated practices. It aligns teams to current business strategy, but it also opens the door to rebuild with clarity and intention.
For those of us who lead these projects, the redesign is more than a task—it’s a digital transformation. And every time I step into one, I’m reminded that complexity isn’t the problem. Lack of clarity often is. Decisions are hard to make from the inside. The job is to bring that clarity back, layer by layer and to look from the outside perspective but follow internal needs.
Goran Paun is the creative director of ArtVersion.