Internet Retailer recently held a Web Design conference for advertisers, focused not only on the aesthetics of web design but also on Website Usability. They did a pre-conference survey and asked the top 500 retailers if they're doing multivariate or A/B testing, and amazingly, 76.7% of retailers surveyed don't conduct these types of tests!
This is incredible if you think about it and an area in which I think the “little guys” can become competitive against the online branding giants, with just a little bit of work.
Before engaging into any type of redesign project, you should be armed with some ammunition of what NOT to do. This data can only come from one source, your customers! Most businesses, stores, restaurants, etc., offer comment cards and truly seek this feedback. Other retailers have gone even further by offering incentives (discounts, coupons, sweepstakes) for customers to complete service surveys on their experiences.
So, with advertisers going to great lengths (and expense) to obtain this type of data in their stores from their guests, why wouldn’t they do the same online? The results above from the top 500 retailers prove otherwise. With some relatively easy and cost-effective (free) multivariate and a/b testing programs available – let your website visitors show you what should stay and what needs to be changed on the site.
Although it’s important to have a website that is pleasing to the eye, it does not mean that users will be able to navigate it easily. By using analytics data and implementing testing, you will truly know when something is working or when it is not. There are no failures with multivariate tests because even those experiments which don’t work have proven successful in terms of knowledge gained.
Before you engage in a website redesign, make sure you have the answers you need to determine what has worked on the site so far and what should be modified or replaced. Without this roadmap, your new site may be lost!