A Website Redesign Isn't a Rebrand. I Confused the Two.

The site is new. You probably noticed. Or you didn't, which is fine – the site has been live for three days and I've already refreshed the analytics fourteen times, so at least one of us is paying attention.

I want to talk about why it took so long. Not because the timeline is interesting – it isn't, it's humiliating – but because the reason is, and it's the same reason I watch founders avoid their own websites for eighteen months while telling themselves they're waiting for the right moment to "do it properly."

There is no right moment. There's only the version you keep not shipping while your current site quietly tells everyone you stopped caring in 2022. Mine was doing exactly that. I was aware. I did nothing.

For the better part of two years, I treated dnsk.work like a client I kept deprioritising. Every few weeks: open Figma, move some things around, decide the navigation felt heavy, shuffle the portfolio grid, make a note that the hero copy needed work, close the file, go back to actual paid work. The client work paid. The Figma file judged silently from the Recent Files panel like a passive-aggressive flatmate.

The symptoms I'd diagnosed: layout too cluttered, hero copy not punchy enough, services page trying to explain everything and landing on nothing. Classic surface-level complaints, all of them correct, none of them the actual problem.

What I'd failed to notice – or noticed, filed under "deal with later," and successfully avoided for 22 months – was that the copy underneath all those layout problems was saying something irreproachably vague.

"Senior UX and product design partner for SaaS teams."

Sure. Technically accurate. Also the approximate equivalent of describing yourself as "a person who does things for money." Correct in the way that's worse than being wrong. Indistinguishable from four hundred other designer websites I could pull up right now. I won't. You've seen them. They all have the same Inter font, a hero section with 60% opacity on something, and a stock photo of a Macbook that isn't theirs.

Here's the thing about visual redesigns: they feel like progress. Decisions are being made. Pixels are moving. The Figma file looks different today than yesterday. That counts as work, technically.

Except when the positioning is broken, no amount of visual tidying fixes it. You just end up with a very well-organised website that still doesn't explain why someone should hire you specifically, instead of the designer two tabs over who also does "senior UX for SaaS teams" and has a very similar sans-serif.

I redesigned the homepage layout four times. Four complete versions, each cleaner than the last, each quietly carrying the same broken positioning underneath it like a structural problem you keep painting over. The fourth version was genuinely good-looking. It still said nothing. I nearly shipped it twice. (I didn't. You're welcome.)

The fifth version I started differently. Closed Figma. Opened a blank document. Wrote, in plain language, what I actually do now versus what I was doing three years ago. Not for the website – just for myself, to see if I could say it without immediately softening it into something more hireable-sounding.

First attempt: four paragraphs, three hedge words per sentence, completely unusable. Second attempt: better, still one foot out the door. Third attempt: I deleted everything that started with "also" or "depending on" and what was left was two sentences that actually meant something.

Design is apparently easier than honesty. Who knew. (Everyone. Everyone knew.)

I'd been using "redesign" and "rebrand" interchangeably for two years. They're not the same thing. Not even close.

A redesign is a visual problem. The site looks dated, the hierarchy has drifted, the portfolio section still features work from a client you'd rather not mention. Fixable. Finite. A Figma problem with a Figma solution.

A rebrand is a positioning problem. It's what happens when what you actually do has shifted – quietly, across two or three years of real work and real clients – and the way you talk about yourself hasn't kept up. The words on the site are still describing a previous version of the business. Accurate once. Subtly wrong now. Not wrong enough to feel urgent. Wrong enough that every discovery call involves a small correction that you've started making automatically, like a tic.

My problem was the second one. I kept reaching for the first solution.

Somewhere between a 14-month enterprise engagement with Deutsche Telekom and a run of SaaS products where the brief said "we need a redesign" and the actual problem was "your onboarding flow is why nobody activates," I'd become very specifically someone who finds the thing quietly costing you users and fixes that – not a generalist who'll do most things if the brief is interesting.

Those are different people. They have different websites. Mine still described the older one.

The redesign often starts with not knowing who you are designing for.

The visual changes took about a week once the words were right. I'm going to let that sit there for a moment.

• The homepage headline went through eleven versions. Eleven. The one that made the cut is four words shorter than the first and says approximately three times more. This is not a coincidence. This is just what happens when you stop trying to appeal to everyone and start describing what you actually do for the people who actually need it. Turns out those are not the same exercise.

• The services page no longer lists every possible engagement model, every project type, every "it depends" scenario I might theoretically take on one day if the stars align. That approach – covering everything just in case – is how you write copy that converts nobody, because nobody reads a thirty-page menu and thinks "perfect, exactly what I needed." They think "I'll come back to this" and then they don't.

• The about page stopped being a credential inventory. Deutsche Telekom is still there because it's relevant. IQVIA is still there for the same reason. What's gone is the paragraph that read like a LinkedIn summary I'd written in a hurry and never updated, because if I wanted to sound like a LinkedIn summary I'd just send people to LinkedIn and save us both the click.

• The work section now leads with the cases I actually want to lead with. Which sounds obvious. It wasn't, for a while, because removing the old ones felt like admitting things had changed. They had. I should have done it eight months earlier. (Classic.)

The reason I stalled wasn't really a design problem. It wasn't a copywriting problem. It was a clarity problem, which is a polite way of saying I was avoiding being specific because being specific means closing some doors, and I wasn't ready to close them yet.

Every time I sat down to write the homepage, I'd hedge. Add a qualifier. Widen the scope just slightly, just in case. Every hedge made the copy vaguer. Every time the copy got vaguer, I'd decide the layout must be the issue and open Figma again.

This is avoidance dressed up as a design process. I've watched clients do this exact thing with their own products – adding features instead of fixing the core, redesigning the UI instead of rewriting the onboarding, adjusting the colour palette instead of admitting the pricing model is broken.

I have, on multiple occasions, said the words "this is a positioning problem, not a design problem" to a founder who did not want to hear it.

I then spent two years doing exactly that to my own website.

I'm a delight to work with, clearly.

Before you brief for a redesign: how to write a design RFP that filters for the right partner.

The new site is not perfect. It's just honest. Which, for someone who tells clients that clarity beats aesthetics every single time, really was the minimum. A low bar. Took two years to clear it.

If it still confuses you – what I do, who it's for, whether any of this is relevant to what you're dealing with – the contact page is right there. I can take honest feedback. I've been practising on myself.

Thinking about a website redesign? Here is how we approach SaaS website design — starting with what is actually broken, not the logo.