The network for creativity
Join 1.25M professional creatives like you
Connect with clients, get discovered, and run your business 100% commission-free
Creatives on Contra have earned over $150M and we are just getting started
Back to feedPost
What I Actually Look at Before I Recommend a Website Redesign
Before I tell anyone to rebuild their website, I try to talk them out of it.
Not because redesigns are bad. Because most of the time, the thing costing them money isn't the design at all. And a rebuild that fixes the wrong problem is just an expensive way to stay exactly where you were.
So before I recommend anything, I look at five things.
The data first, opinions second.
I don't start with how the site looks. I start with what people do on it. Where they land, where they drop off, what they ignore, how far they scroll. The analytics usually point straight at the real problem, and it's rarely "the whole thing needs to go."
Traffic problem or conversion problem.
These need opposite solutions. If people are arriving and leaving, no redesign fixes that on its own; the message or the offer is off. If nobody's arriving at all, that's SEO and visibility, not layout. Diagnose this wrong and you waste the entire budget.
The message before the design.
Most underperforming sites don't have a design problem. They have a clarity problem. If the headline talks about the company instead of the visitor's problem, people leave in seconds, and a prettier version of an unclear page just fails more attractively. Usually a copy fix, not a rebuild.
Speed and mobile reality.
I check how fast the site loads on a mid-range phone on mobile data, not on a fast laptop. A slow load kills conversions and rankings at the same time, and it's often a performance fix rather than a full redesign.
Where trust is, and isn't.
I look at whether proof shows up where doubt lives, right next to the ask, or whether it's buried on a page nobody visits. Moving social proof is a small change with an outsized effect. Again, not a rebuild.
After all of that, one of two things is true. Either I've found targeted changes that lift results without a full rebuild, which saves the client real money. Or the site does need a redesign, and now we go in knowing exactly what to fix instead of guessing.
A redesign should be a conclusion, not a starting assumption.
The goal was never a new website.
It was more revenue. Sometimes those are the same project. Often they're not.
If your site is underperforming and you're not sure whether you need a rebuild or just a few targeted fixes, send me a message. I'll tell you which one it actually is.
The network for creativity
Join 1.25M professional creatives like you
Connect with clients, get discovered, and run your business 100% commission-free
Creatives on Contra have earned over $150M and we are just getting started
Related posts
So proud to finally share this one 🎨
Drawing America was one of our biggest Kajabi projects – a 4+ month rebuild and migration of an entire art education platform from WordPress to Kajabi.
We started with the sitemap and platform structure, created a visual guide, designed the experience in Figma, and then moved into custom Kajabi development.
One of the most interesting challenges was building a custom filtering system for 50+ courses across recorded, live online, and in-person formats. On top of the website, we also worked on course migration, Kajabi offers, events, community, automations, and migrating 9,000+ student contacts.
Definitely one of those projects where Kajabi became much more than a course platform 🙌
Very impressive. I’d love to help out if you need help with AI automation or an extra dev support in the future
It's been 3 years since I started UI Flip and its website was the first one I built in Framer.
It was far from great and an overhaul was way overdue.
I started a new version from scratch earlier this year and it has been on-hold for months, as I wanted to polish it just a little bit more, but this morning I said "f*ck it, just ship it" and hit that "publish" button 🚀
i’ve spent years designing websites with whatever tools were available
now i’m slowly getting used to the fact that i can also make the tool
for this project i wanted to explore more directions for the brand
collecting references or creating each direction from scratch would already become a project of its own
so i built a small generator and started playing with it
each output gives me something to react to
keep this
change that
try one more
and then one combination suddenly feels like a direction worth following
it’s still rough and i keep finding small things to fix
but honestly this is the part i’m enjoying most
doing the design and building the tool that helps me explore different visual languages at the same time
Making the tool instead of just using it is such an underrated superpower for designers. Once you can generate directions to react to, exploration stops feeling precious and starts being fun. This is the good kind of lazy.
