You probably don’t need a redesign

In short: Most requests that begin with “we need a redesign” turn out to be content and structure problems in a design costume. A rebuild resets the clock, arrives as one large bill, and hands back the same problems a few years later. The cheaper path is unglamorous: fix what the site says, fix how it’s arranged, make it fast, then keep it in repair. That last part is the one nobody budgets for.

The message is always about the look

The note usually reads the same way. The site feels dated. Inquiries have slowed. A competitor launched something sharper last month. Can we talk about a redesign?

It’s a fair request, and the instinct behind it is sound. Something is wrong, and the website is where it shows. But by the time that message gets written, a symptom has been spotted and the most expensive cure prescribed in the same breath.

This is an awkward argument for me to make. Redesigns are a good part of what my team at Dthree Digital builds. Talking someone out of one is talking myself out of the larger invoice. I keep doing it anyway, because the alternative is watching a client spend a year’s marketing budget to arrive back where they started, with nicer typography.

What tends to be broken

When I go looking, the same four things turn up, in roughly this order.

The content hasn’t been touched since launch. Services that changed two years ago, a team page listing people who have moved on, an About page written to fill a space rather than answer a question.

The structure follows the organization instead of the visitor. Menus built from departments and internal vocabulary, so the thing most people came for sits three clicks deep, filed under a word only the company uses.

It’s slow where it counts. Fine on office fiber, painful on a mid-range Android on mobile data, which in the Philippines describes most of the people you are trying to reach.

And nobody can edit it. The CMS is either intimidating or genuinely broken, so every small change becomes a request to an agency, and small changes quietly stop happening.

A redesign changes what a site looks like. Not one of those four is a looks problem.

You can redesign a page that says nothing. It will just say nothing faster.

A test before you sign anything

Three questions, and answering them honestly takes about an afternoon.

When did anyone last change a sentence on the homepage? If the answer is launch day, the site isn’t dated so much as unattended, and a new one will get there the same way.

Can you name the three things a first-time visitor is trying to do, in order? If that takes more than a minute, the trouble is structural, and structure survives a redesign unless somebody deliberately changes it.

What does the site feel like on a phone, on data, away from your desk? Not the score from a testing tool. The actual wait.

If those answers are uncomfortable, a redesign will cover them beautifully for a while. Then they surface again, because none of them were touched.

When a rebuild is the honest answer

Sometimes it genuinely is, and pretending otherwise would be its own kind of sales pitch.

Rebuild when the brand has changed, not merely aged. Rebuild when the platform is a dead end: an abandoned theme, a page builder that lost its maintainer, a stack where routine changes cost more than they should. Rebuild when the templates themselves are the accessibility problem. Rebuild when what the business sells is no longer what the site was built to sell.

Those are structural, you can’t write your way out of them, and patching around them for three years costs more than starting cleanly. It’s the same distinction I’d draw about platform migrations: the question isn’t whether the new thing is better, it’s whether the old one is the real constraint.

Where the money should go

Here is the part that changes the decision, and it’s arithmetic.

A redesign is a lump sum. It lands once, resets the clock, and buys a site that is current on the day it launches and never again. Three or four years later the same conversation starts, for the same reasons, at roughly the same price.

The same money spread as upkeep behaves completely differently. Content gets corrected while it’s still wrong instead of at the next rebuild. Slow pages get attention when someone notices them. The structure moves as the business moves. The CMS stays something the team can actually use, so the site keeps absorbing small improvements rather than accumulating small failures. Nobody has to relearn anything, because nothing was thrown away.

That is the whole case for maintenance, and it is the least exciting thing anyone will ever pitch you. There’s no launch announcement at the end of it. It just means that in year three the site is better than it was in year one, which is not what usually happens.

Most sites don’t get worse because they were designed badly. They get worse because nobody was responsible for them after launch.

Before you sign off on a rebuild

Try this first. Write down everything you would fix if a redesign weren’t allowed. Just the list.

If it’s short and cosmetic, go build it with a clear conscience. If it’s long and nearly all of it is content and structure, you have just found the actual project, and it costs a fraction of the one you were about to approve. That steady work is what my team’s support plans exist to do. If you can’t tell which list you’re holding, that’s a conversation I’m glad to have, including the version where the answer turns out to be yes, rebuild it.

Quick questions, quick answers

How often should a website be redesigned?

There is no clock. Four to six years is a common rhythm, but the trigger should be a real change in the business, the brand, or the platform, not an anniversary. A site that is kept up can run much longer.

What does upkeep actually cover?

Content edits, speed, broken links, platform and plugin updates, structural fixes, and someone occasionally looking at what visitors do and adjusting the site accordingly. That last one is the part that compounds.

Won’t a redesign help our search rankings?

Not on its own, and a careless one hurts: URLs change, internal links break, and years of accumulated signals go with them. Rankings improve when the content answers real questions and the structure is clean, neither of which requires rebuilding anything. I wrote about the diagnosis side of this in Your traffic is down and it’s not your fault.