BLOG

on Conversion Rate Optimization

See How a Quick 2-Minute Test Increased eCommerce Conversions by 36.54%

Do you want to have a conversion-centered design?

Well, who doesn’t? Wrong question.

Let me put it another way – how much do you know about designing for better conversions?

If you are awesome, you will know that the page elements that have most impact on the decision making of the prospects must get the attention they deserve. Often, this means increasing their size, giving them contrasting colors, or changing their placement on the page.

The point is, conversion-centered design should ensure that page elements which influence prospects’ decisions the most are easy to spot when visitors are scanning the page.

In this case study, we show how Royal Discount used Visual Website Optimizer to achieve valuable testing wins by making important information stand out.

About the Company

Royal DiscountsRoyal Discount offers computer hardware and software from recognized manufacturers like Microsoft, Adobe, Monster Cable and Symantec at discounted prices. The company has two main competitive advantages — 1) their top-notch customer service 2) extremely low prices. They hired Fruition.net to improve the conversion rate of their website.

Test Hypothesis

As is clear, pricing plays a crucial role in the decision-making of their target audience. So, the company first started out with a product page A/B test, where they added crossed out MSRP (Manufacturer’s Suggested Retail Price). This gave a little boost to their conversion rate.

Next, they hypothesized that price information is getting lost in the remaining text on the page. Even though their call-to-action was standing out for visitors, they were probably missing out one of the key reasons why they should make the purchase from the Royal Discount website.

So, the hypothesis was that increasing the emphasis on pricing will increase conversions. “Add to cart” and the Final purchase were the two goals they were tracking. They made the change on their Microsoft Office 2013 product page to see how it works out.

This is their Original page:

Royal Discount's original product page

Fruition.net made a quick A/B test in Visual Website Optimizer, where they increased font size of the sale price from 15 px to 20 px and made it bold. The price of MSRP remained the same though. And this was their Challenger page after the change:

Royal Discount's Challenger page

Result

The test ran for more than one month. The Challenger page pushed 36.54% more visitors into the conversion funnel for their “Add to cart” goal. And this increased their revenue by 10.21%, although the statistical significance for this goal hasn’t been achieved yet.

You can compare the two versions in the image below:

Royal Discount's Comparison Image

When buying a popular software product like Microsoft Office most people are not too concerned about great-looking pictures or product descriptions. Authenticity and pricing are their main concerns. By making a simple change, Royal Discount made it easy for their customers to make the decision to buy the product.

Suggest Some Tests for Royal Discount

Fruition.net plans to run some more tests to increase conversions for Royal Discount. What ideas do you have to share with them? Send in your test ideas in the comments section and I’ll definitely convey them forward.

Comments (9)

Leave a Comment
  1. Possible elements to test:
    1) Text colors for the sales price (e.g. green for the sale).

    2) I’d suggest testing “Sales Price” at the new, larger size and “$174.99” at the original, smaller size. (One pricing psychology study at Clark University & the University of Connecticut found that people tend to perceive prices as less when the numbers are smaller).

    3) If you can get authorization to do so, testing a sales value of $169.99 vs. the current $174.99. My hypothesis here is that having so many different digits in the number takes more mental effort to process than if digits were repeated. In addition, studies have shown that 9 is a “charm number” which is associated with value.

    3.1) Testing the removal of the $.99 to make the value easier to mentally process. E.g. testing $174 vs. the current $174.99.

    4) Replace “Add to Cart” with something more direct like “Buy Now” (or perhaps “Download Now” if applicable).

    5) Test the color of the add-to-cart button. If you haven’t already tested green, I think that’s frequently an effective CTA color because green is associated with progress / moving forward / success.

  2. @Richard: Yes, A/B testing (or often Split testing) is very useful to increase online revenues in most of the cases. However, many online businesses don’t know those techniques and lose money.

  3. In UX design basics – some are running, some are just starting to know how to tie shoelaces. And yes, this is the second case.

  4. They should run the same test on another product and combine the results to have significant results. With such a big increase in conversion on such a long period, looks like they could have planned the test better.

  5. I designed this particular test and wanted to jump in and respond.

    @B. McKenzie, thanks for the ideas. This was actually a multi-variate test that experimented with adjusting the size of the retail price as well. The combo on the winner yielded the best results. We’re also currently running a test on the checkout buttons.

    @Goofy, the site design and UX leaves much to be desired, but we work with what we’re given. My team was tasked with leveraging as much as possible without a redesign.

    @Benoit, this test used the wildcard option and it ran on all the Microsoft Office products they sell which is over 100 different SKUs.

  6. Thanks for the reply Todd. I’m surprised that you haven’t reached statistical significance yet then, especially with the high difference in conversion rates.

  7. We reached a 98% confidence rate in the add to cart goal and an 83% confidence rate in the purchase goal. Due to a few variables I’m not at liberty to discuss, the add to cart goal was more important for this test. Given a little more time I’m pretty sure we would have seen the confidence level of the purchase goal reach the upper 90’s as well.

Leave Comment

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes : <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>

Contact Us / Login

Product
Resources Home