Accelerating Business Growth Through Website Testing

Let’s just get this out of the way: it’s never easy admitting that you don’t have all the answers. Despite the technological advances we’ve made, so many of us have a difficult time accepting that we don’t know as much as we think. That’s where testing comes into play.

Traditionally, many ecommerce sites have kept online testing at its simplest level. Major site modifications are still made solely on the ‘vision’ of top members of the organization. While these changes are typically based on some research, without actually testing them against the original, it is unknown if changes had a positive impact.

More progressive corporations, however, reject this approach and instead drive their marketing and product strategies through controlled website experimentation. Companies like Microsoft, Amazon, Netflix, Priceline have all allowed their consumers to navigate the direction of their website instead of relying purely on the ideas of their marketing teams.

These organizations run thousands of tests yearly to improve upon winning variations and disregard any losing variations. To them no one’s opinion is more important than their users. This has proven a rewarding tactic to say the least:

  • Priceline.com’s utilization of A/B testing to determine what drives users to convert “has resulted in conversion levels 2-3x the industry average.”
  • Microsoft’s Bing saw its share of U.S. searches conducted on personal computers rise from 8% up to 23%, in just three years.

So, how can companies without digital roots create a culture of experimentation within their organizations? We’ve gathered the 3 key steps to getting started:

Accept that failure is part of growing your business

While many organizations have an in-house CRO expert, most product decisions and website changes are still made at board level. Rather than testing new features and changes, they’re quickly rolled out to users and deemed a “victory.” When success is only measured by the number of products and features released, and not their actual impact, failures and shortcomings are often ignored.

Organizations should strive to know the effect, positive or negative, of every change they make to their website. It’s only through trial and error that organizations finally start inventing successful products and no company understands this better than Amazon. Don’t believe me? Their CEO, Jeff Bezos, touched  on this exact point in his letter to shareholders:

“One area where I think we are especially distinctive is failure. I believe we are the best place in the world to fail (we have plenty of practice!), and failure and invention are inseparable twins. To invent you have to experiment, and if you know in advance that it’s going to work, it’s not an experiment. Most large organizations embrace the idea of invention, but are not willing to suffer the string of failed experiments necessary to get there.”

By accepting that failure is unavoidable, you empower your team and provide them with the platform they need to create innovative products. Controlled experimentation is one of the best ways to determine the success or failure of your teams ideas, so it’s imperative that every level of the business is invested in testing.

That leads me to my next point…

Even the best ideas are cheap. Test everything!

Everyone from the CEO down to the newest associate has an opinion about how a product change could impact conversion. Some base it on their gut feelings, while others observe data and build a strong hypothesis based on their findings. Yet when tested, many ideas still negatively impact or fail to improve metrics. In fact, a 30% success rate is considered high. It goes to show that no matter how much you think you know about your customers, their way of thinking will constantly amaze you.

A great example of this is a test run by JoAnn.com, the E-tailing arm of $1.7 billion arts-and-crafts retailer JoAnn Stores Inc. The experiment aimed to find out how they could sell more sewing machines. They were surprised to learn that the winning variation was one in which people who bought two sewing machines saved 10%. Even their CFO was perplexed as to why anyone would need two machines.

After performing their own research, JoAnn.com concluded that the online arts-and-crafts community was a tightly knit group, and users were coming together to take advantage of the deal. Experiments such as these not only increase revenue, they also shape future business decisions.

While experimentation is generally associated with increasing revenue, A/B testing can also help mitigate revenue loss due to unknown bugs in products. When Amazon wanted to release a new version of a particular feature on their app, tests showed that the new variation caused a 2% revenue loss. When it was determined that the issue was due to an unknown bug and then corrected, the team ran another test. Again, the new variation failed. The new feature shipped after five total iterations.

This is exactly why companies like Amazon and Netflix rigorously A/B test every change before it becomes the default user experience. To them, any change is “too risky to roll out without extensive A/Btesting.”

Let customers decide the direction of your website

Corporate culture is notorious for responding slowly to user needs and producing work that has incremental value. This is mainly due to an archaic business process that attempts to outline large scopes of work with lengthy production schedules. But if they want to be agile, organizations need to focus on features that add user benefits and increase revenue now. The only way to do that is to understand what their users respond to.

Organizations can learn what’s truly important to their users by running website experiments. If your hypothesis is correct and users are sensitive to a particular change, you could make the case that more resources should be allocated to optimizing that portion of the website. With that kind of data available, you can be confident that you’re prioritizing funds toward a potentially crucial step forward in optimizing your website. However, if users have a negative (or worse, indifferent) reaction, you’ll need to reevaluate your hypothesis.

Most importantly, consumers help selectively optimize websites by expressing which traits to remove and which to build upon. Stuart Frisby, the director of design at Booking.com recently admitted that “Booking.com doesn’t look the way it does because I decided if it should look that way and if I had, then it wouldn’t look that way, I can promise you that. It looks that way because over eight years, customers have told us that this is the product that they want, in small measurable incremental steps.”

If you don’t know where to start, below is a list of elements I suggest you experiment with:

5 Tips To Help You Start Website Testing

  1. Search – Often overlooked but extremely valuable to visitors. The search functionality, design, placement and workings (autocomplete, more specific suggestions, etc) is a great place to start testing. Amazon is a great example for how search functionality should work. 
  2. Shopping cart – Is one of the most crucial steps of the conversion funnel. Visitors can easily get anxious and abandon their order so test how you present messaging such as, free shipping eligibility, expected delivery date and how sales prices are displayed. 
  3. Product category page – Category pages are a prime testing opportunity. These pages show the breadth and range of your products and need to convey a lot of information.The correct page layout and product description is key to improving conversion. 
  4. Sign up pages – Users get impatient. Progress indicators during sensitive conversion steps are a great way of reducing anxiety and restlessness and improving conversion rates. Dropbox.com subtle use of this is a great example. 
  5. Trust messaging – How and where you display product reviews, user comments, return and exchange policies, customer support number can improve or hurt conversion. It’s key to experiment with different design variations and placement to find what your users best respond too. 

As online shopping revenue continues to grow year on year, companies that ignore full-scale experimentation and continue to make unscientific decisions about their website’s will make the switch to a test-driven strategy. In the age of digital, there is no choice in the matter: either businesses listen to and react to their users or a global competitor will. Those already using testing to drive their conversions? They’ll see their conversions skyrocket while adapting and improving the way they test for increased company growth.