Something To Think About Before You Redesign

We work in a redesign obsessed industry. I make regular rounds across the Internet Retailer 500, and one thing is constant - change. A 2009 Internet Retailer survey found more than 60% of web retailers redesigned all or part of their websites that year, and 67% planned to redesign in 2010 -- and that despite a weakened economy. We know site design plays a major part in ecommerce business performance, but design projects can cost thousands, even hundreds of thousands of dollars.

Retailers redesign in hopes to boost sales

Top reasons given by survey respondents for redesign were to boost sales and conversion, attract new shoppers and improve customer service. To improve “shopability,” many redesigns focused on reorganizing content throughout the site (44%), improving navigation and search pages (31%), and speeding up the checkout process (28%).

While in theory this is all good, there is risk in improving a site too much all at once. When you redesign processes that customers and frequent visitors have learned, even if they're less-than-optimal processes, you introduce a new learning curve. This can, at least in the short term, negatively impact your site performance. Or, the design could simply underperform across all visitors, and performance will suffer until the site is truly "improved."

Modern love?

A concerning stat: 34% of retailers cited outdated design and graphics on home, category and product pages as the biggest drawback of their current designs. This indicates the motivation to redesign is based on gut-feel that sales would increase if the site looked more modern, rather than data or customer input. Of the 75% that claim to test their websites, only 36% use A/B testing and 19% use multivariate testing.

Sadly, many redesigns, new processes and features are scheduled to launch "just in time for the holidays" untested - just in time to kill sales?

Avoiding redesign pitfalls

1. Survey existing customers / user test to discover what could be improved on you site, use that to shape your redesign plan - rather than an internal brainstorm, or current design trends. (Don't fix what ain't broke)

2. Make small, incremental changes rather than one massive UI makeover

3. Roll out changes in stages, first to “new” visitors (uncookied), then to returning visitors and finally, returning customers (depending on the feasibility with your testing tool and segmentation)

4. Don't launch major revamps or new features just before your busy season, give yourself time to work out the kinks and measure conversion (or recover from short-term dips) during the "off-season"

5. For major changes, educating returning customers on the new features / improvements can help (site tours or tutorials)

For example, Amazon:

Wanna learn more about site optimization testing? We're teaming up with Marketing Sherpa for our August webinar Putting Your Best Site Forward: Making the Case for Consistent Testing on Enterprise Ecommerce Sites on Wednesday, August 24, 2011 from
9:00 to 9:45am PDT.