- The Conversion Ledger
- Posts
- The Hidden Cost of the Redesign
The Hidden Cost of the Redesign
They don’t tank conversions because they are ugly.
Brian Schmitt
September 07, 2025
(Sending it late today. Been one of those weeks.)
You’d be shocked how many redesigns destroy revenue.
On the surface, a redesign feels like a strategic play. Modern UX, fresh branding, and maybe even faster load times. But in the CRO world, we’ve seen it too many times: a client calls us three months post-launch with the same panicked script:
“Our conversion rate dropped… and we don’t know why.”
They assume it’s a one-variable problem. It’s not.
What they’ve really done is torch years of optimization. Every button placement, every headline, every buy box that had been iteratively improved through A/B testing… gone.
The Myth of the Redesign
Here’s what most ecomm teams believe:
“Our current site is outdated. Let’s start from scratch.”
“If we rebuild with the user in mind, conversions will follow.”
“We’ll test after launch.”
But here’s what actually happens:
You swap proven UX for untested assumptions.
You change 47 variables at once and lose the ability to isolate causes.
Your funnel breaks in ways your team anticipate and your inventory becomes a problem.
Let’s be clear: redesigns are not inherently bad. But unvalidated redesigns are. Especially when they remove high-performing elements without realizing they were working.
Win of the Week:
We worked with a mid-size DTC brand that wanted to borrow a “personalized homepage” strategy from a larger competitor.
They saw the competitor crushing it and thought, “If it worked for them, it’ll work for us.”
They tested a Chewy-style homepage layout on 42,477 visitors:
Personalized modules for auto-replenish
Rewards reminders
Product review prompts
Mobile saw minor wins. But desktop? Engagement tanked. The layout distracted more than it converted, especially for new users unfamiliar with the brand.
Takeaway: Copying a successful design treatment without context can backfire. What works for a loyalty-heavy brand doesn’t always apply to a first-time buyer journey
Quote of the week:
We must learn what customers really want, not what they say they want or what we think they should want.”
If you’re considering a redesign:
Start with your highest-converting pages. What’s working? Why?
Run usability research before any designs are started.
Run user testing with clickable prototypes.
Isolate variables. Consider a rolling redesign, where you test one major section at a time.
Measure everything. If you’re changing headlines, hero layout, and CTAs simultaneously, you’re taking a bigger risk. Understand that going into the test.
Redesigns aren’t wrong. But they should be treated like a high-stakes test, not a branding exercise.
Looking forward,
How valuable was this week's newsletter? |
P.S. Ready to grow revenue without having to grow traffic? Let’s talk.