If you love to risk destroying your conversion rate in mere hours PLUS spend a bunch of money doing so, make sure you do this ONE trick: launch a brand new redesigned website overnight.
Sarcasm aside, redesigning and launching a new version of your site without testing is one of the riskiest things that a brand can do.
For some reason, when people start to split-test and perform CRO, after a few weeks they decide to “redo their entire site”.
I get it. They’re excited about converting better, realizing how many improvements they had to make, dislike the look of their current site, and figure that they should just knock it all out at once.
However, it’s pretty much the worst thing you could do.
I’m going to get mean for a second, but I promise it’s worth it.
Site redesigns are beloooved by CMOs who aren’t executing day-to-day, who aren’t seeing day-to-day data, and who see a competitor’s website and get jealous of the design all while having ZERO idea of how it’s converting. It’s something they can go to the rest of the team/board and say “look how pretty it is!”
Pretty doesn’t mean it converts. Pretty CAN convert… but just because it looks good doesn’t mean it will.
Just go look at any venture backed SaaS company’s website versus their cost per acquisition and you’ll see what I’m saying.
Alright I’m done being mean.
Let’s go through this logically:
If you’ve been running split-tests, you know that about 50% of the time they have no benefit, or they actually worsen your conversion rate
Often, what we *think* is going to work doesn’t – meaning that our preconceived notions are often wrong
Pure design changes often have much lower impact on conversion than we think
Now, let’s keep going with some truths about site redesigns:
- Website redesigns cost a lot
- They take FOREVER (always twice as long as you think)
You’ll want to rip your hair out with endless revisions and project management
Put these two together, and you have something that will cost a lot of money, take a ton of time, and have a large chance of actually harming conversion rate.
So even though doing a site redesign can “feel” good, the risk almost never is worth the reward.
Imagine you spend $30k for a new site, launch it, and your conversion rate goes down by 40%.
I wish I were exaggerating, but I’ve seen it happen.
Obviously, it’s not all black and white. Sometimes a full site redesign may be in order if branding is changing, if your business got acquired, if you’re changing markets or product suite etc. But even in those cases, I’d try my absolute hardest to not just pull the old site down and put a new site up.
So what’s the solution?
You want to test your way to a new site.
Let’s say you HATE your current site design. Fair enough.
What you should do instead is redesign your page that gets the most traffic and split-test it. If the new version wins, then keep redesigning and testing page by page.
Now, you’ll have data-backed proof to support your hypothesis.
And if the redesign loses? Then you’ve just saved yourself a whole lot of money (and wasted conversions).
We’re trying to mitigate risk – and the risk that a brand new site kills your conversion rate is simply too high to justify.
