Ellen Julian

Don't Fall in Love with Your Site RedesignThe blessed day finally arrives. After months of labor selecting images and messages, the design team is ready to deliver a simpler version of your website. Excitement abounds as designers reveal gorgeous images uncluttered by pesky text. Yes, millions of visitors are about to meet your new baby!!

But wait, has anyone tested the redesign? Of course, the design team studied the site through the eyes of a potential buyer and ran a bunch of focus groups. Everyone loves the new look. It’s time to push it out and start selling!

Not so fast, the Optimization Leader says. Let’s get some data before all of our visitors try the new site.

Fast? We’ve been working on this for a long time, says the Design Lead. We know this look works for site X, and they’re outselling everyone. This is an innovative design, and we can’t wait any longer. We’re ready to go and we have to make our numbers for Q2. Let’s do this!

The design team launches the new site and within weeks, there’s a steep reduction in conversions site wide. If only they had done more testing to really know what visitors want instead of assuming based on limited data.

Does this scenario sound familiar?

Companies frequently fall into this trap because they are blinded by affection for a newer, sleeker design. Paul Terry, Senior Optimization Consultant at SiteSpect and former Web Optimization Analyst at PRIMEDIA, advises companies to avoid letting their attachment to new designs run amok. “Simpler isn’t always better,” says Terry. “Sometimes removing relevant product information can have the opposite of the desired effect; less information may depress conversions. Understanding what major design changes mean to conversion helps mitigate risk.”

This advice rang true for Wayfair. The retailer had high hopes for a complete page redesign that failed to produce gains and led to a drop in the site’s conversion rate. “One of the biggest challenges to testing site changes is realizing that most tests aren’t winners,” says Jeff Hannan, Wayfair LLC’s Senior Manager, Site Testing and Customer Analytics. “The nature of site changes is that most tests have neutral or losing results. When we don’t have a winner that means we have to regroup and push forward.”

Trulia agrees. “Testing with SiteSpect offers us a way to measure the side-by-side performance of these designs with a much larger sample of users than would be possible in a qualitative study, “ said Dan Voorhies, Senior Director, Data Warehouse & Reporting at Trulia. “SiteSpect works well for traffic management, allocation, and analytics for the larger-scale testing that is needed for major changes in the user experience.”

So take off those rose-colored glasses. Chances are your new baby will deliver increased conversions, but testing is the key to knowing.

Tags: A/B & Multivariate Testing Best Practices