Website redesign: WordPress, Webflow or custom? The honest comparison

A website gets rebuilt every 3 to 5 years on average. Not because of fashion: visitor expectations keep rising, and both Google and AI engines now judge speed and content clarity.

Before launching a redesign, check that you actually fit one of these cases. Rebuilding a site that works, just for the pleasure of a new design, is the most useless expense on the web.

This is the real question behind most redesigns, and the answer is rarely "because the technology is bad". Three problems come up almost every time, and none of them is solved by switching platforms.

The message is unclear. Within five seconds, a visitor must understand what you do, for whom, and what is in it for them. There is no proof. No portfolio, no numbers, no testimonials with a name and a face: the visitor has no reason to believe you.

The site is invisible. If nobody finds you on Google, or through ChatGPT or Perplexity, the best design is worthless. Content matters more than technology here: we explain how in our guide getting visible on ChatGPT and AI engines.

If your site is fast and readable but does not convert, start with the message and the proof. The technical redesign comes after, never instead.

When a WordPress site slows down, the standard answer from hosting companies is to sell you a caching plugin or a bigger server. It works for a while: if your site is less than three years old, with few plugins and a lightweight theme, a serious cleanup (removing unused extensions, compressing images, configuring the cache properly) can be enough and costs €300-800.

The problem is the threshold where patches stop working. A 5-year-old WordPress site typically accumulates 20 to 40 plugins, a theme bloated with options, and fixes layered on by successive vendors. At that point, caching no longer addresses the cause: the site runs too much code to display a simple page.

The honest test: if your site has already had two "performance" interventions and became slow again each time, patches are a money pit. That budget is better invested in a rebuild, on a slimmed-down WordPress if the blog is central, elsewhere if it is a showcase site.

The trap with this question is that almost everyone answering it has skin in the game: Wix recommends Wix, WordPress agencies recommend WordPress, hosting companies sell servers. At Sonho, we sell neither hosting, nor WordPress, nor Wix: we hand-code custom websites, and we will tell you when that is not the right choice for you.

WordPress is the right choice when content is the heart of the site: a large blog, several writers, hundreds of articles. Webflow suits a marketing team that edits its own pages every week and accepts a subscription for as long as the site exists. A custom build (a site hand-coded with modern tools like Next.js, plus an editing interface like Sanity) suits a 5-to-30-page website that needs to be fast, durable and entirely in your name.

Here is the comparison we wish we had found when clients ask us the question:

Comparing quotes on the entry price is a reasoning error. A WordPress site billed at €3,000 commonly costs €8,000 to 12,000 over three years once you count hosting, maintenance, plugin licenses and incidents: an update that breaks the theme, a hacked extension, a form that stops sending.

These costs are not vendor abuse: they are structural. WordPress is an assembly of third-party building blocks that each evolve at their own pace, and someone has to pay to keep the whole thing standing. Webflow shifts that cost into a subscription: more predictable, but never-ending.

A modern custom site flips the equation: no plugins to update, no extension vulnerabilities, no hosting to manage. Once live, it just runs. At Sonho, a hand-coded custom website starts at €2,000 excl. VAT as a fixed price: more expensive than a WordPress theme installed in two days, often cheaper than that same WordPress three years later. And you edit everything yourself through the editing interface, without paying to change a photo.

This is the number one fear, and it is justified: a poorly managed redesign loses 5 to 7% of traffic on average, sometimes much more when URLs change without redirects. Good news: it is a risk of method, not of technology. You can migrate from Wix, WordPress, Squarespace or Webflow while keeping your search rankings, as long as you follow four steps.

First, an audit of the pages that rank: list in Search Console the pages that receive Google traffic and their keywords. Those are the assets to protect. Then, 301 redirects: every old URL permanently redirects to its new equivalent page, page by page, not everything to the homepage.

Third step, preserving the meta tags: the titles and descriptions of pages that perform are carried over, never rewritten at random. Finally, post-launch monitoring: submit the new sitemap in Search Console and watch for errors for four to six weeks. A dip lasting a few days is normal; a drop that persists signals a forgotten redirect.

This is the topic no comparison ever covers, yet it is the first blocker we run into: a company wants to rebuild its site and discovers that the domain name is registered in the agency's name, that nobody has the hosting credentials, and that the agency no longer answers. Getting your own site back then takes months.

Before signing with any vendor (us included), demand written answers to this checklist:

Our position at Sonho is simple and deliberate: the client owns everything. Domain in your name, content yours, credentials with you. If you ever leave, you leave with your site. To check where you stand on ownership and search rankings before a redesign, tell us about your project: the first conversation is exactly for that assessment. Timeline-wise, allow four weeks for a simple website, six weeks on average: the details are in our guide how long it takes to build a website.