5 Businesses That Used A/B Tests to Lift Conversion Rates by up to 216%

Image Source

When I think of A/B tests, I’m immediately taken back to the years of eye exams where the doc asks, “Which is better, 1 or 2?” The patient’s response helps the doctor fine tune a prescription for the laser-sharp eyesight.

A/B tests serve much the same purpose for marketers. A/B testing lets Web designers and marketers fine tune a prescription to build the most effective landing page. Like the vision test, the changes between two landing page elements can be subtle and barely noticeable to the unassuming eye, but make a big difference to the user and conversions. It helps make what works, and what doesn’t crystal clear.

The following five businesses used A/B testing to fine tune the element of their landing pages to improve performance and increase conversions.

1. WriteWork

WriteWork.com was finding that despite a significant amount of organic traffic, they were seeing limited conversions. It must be the landing page. After conducting several customer feedback surveys, they decided to overhaul their landing page completely. The company not only redesigned the site, but changed the site copy and key words to make the language more relevant to their audience.

Page A – WriteWork’s original landing page showed no clear benefits for shoopers who responded to the Call to Action (CTA), reducing conversions

Page B – The new page made highlights benefits to customers and helped to reduce the bounce rate

Conversion Results

The A/B test, which was completed by Website Visual Optimizer, showed the radical redesign was a hit. In fact, engagement more than doubled.

  • The new site and use of the primary CTA increased engagement 144%
  • Ultimately, a follow up test revealed that payments for services rose more than 50%

Testing Service & Full Case Study: Visual Website Optimizer

2. RJ Metrics

Preparing to conduct their very first A/B test, RJ Metrics knew they needed to define goals which were achievable and measurable and lead to greater revenue. Since the company’s website doesn’t directly generate revenue they chose to test their free demo offer, the one measurable action that brought the consumer the closest to the final sales process.

Page A – RJ Metrics original landing page was busy with a video explaining the product. The CTA is small and does not provide any information as to what the free trial entails

Page B – The revamped page removed the busy video, significantly increased the size of the CTA and specifically defines the benefits to the customer who opts for the demo

Conversion Results

Using Google Website Optimizer, RJ Metric’s A/B test showed the new site to be much more effective than the original.

In fact, conversions rose 190%.

The best part, the video from the original page didn’t go to waste; it’s now being used on another page within the RJ Metric’s website.

Testing Service & Full Case Study: Google Website Optimizer

3. Obama08

During President Barak Obama’s 2008 presidential campaign, his campaign committee wanted to optimize every aspect of his campaign. The thought was that the right web design potentially meant the difference between raising $10,000 and $10,000,000 in campaign funds. The team looked at 24 different variations of the splash page using a mix of images and CTA buttons to determine which combination produced the greatest results.

Page A – The Obama team tested several combinations of visuals and CTAs to identify the most effective pair

Click for full size image

Page B – The winning combination engaged emotion with the image of the Obama family in conjunction with the Learn More button

Click for full size image

Conversion Results

With more than 300,000 visitors to the splash page during the testing period, each variation was seen by more than 13,000 people.

  • The winning combination saw a 40% increase in sign-up rates.
  • Though the number may not be as lucrative as some of the other examples examined for this article, the overall perceived success is staggering. It’s estimated that the increase in conversions resulted in an additional 2.8 million email addresses which ultimately led to $60 million in donations.

Testing Service & Full Case Study: Optimizely

4. Kwik Fit

The business of Kwik Fit, a U.K.-based insurance intermediary, is greatly dependent upon its digital presence. Because of this, the company wanted to find a way to increase conversions from visitors to the landing page for its highest profiting product line. They decided to conduct an A/B test that looked at page-layout, content and design on its affiliate landing pages.

Page A – The original affiliate landing page was text heavy with many of the unique benefits falling below the fold

Page B – The winning design optimized the sites content and helped Kwik Fit meet its goal to increase conversions

Conversion Results

Utilizing Maxymiser’s Multivariate testing software, KwikFit recognized their original design was too busy and lacked a prominent CTA.

The winning design, which included less text, unique selling points located above the fold and a more prominent CTA, improved registration rates by 12% and increased the number of confirmed purchases by 78%.

Testing Service & Full Case Study: Maxymiser

5. Buy Real Twitter Followers

Taking advantage of a growing Twitter advertising market, Buy Real Twitter Followers works to help advertisers increase their clients’ Twitter presence. But they needed to increase their own sales as well. So, they devised an A/B test that would provide feedback on the addition of a video to their home page.

Page A – The original Buy Real Twitter Followers homepage is text heavy which many visitors would simply skim through

Click for full size image

Page B – A simple addition of a video maximizes the content by drawing attention to the product information

Click for full size image

Conversion Results

The minor change of adding the video to the landing page helped them increase their sales by an astonishing 216%.

Because visitors generally only scan through text-heavy material, the video draws more attention, providing a better understanding of the product. Despite the stellar results, their current home page is missing the video, but it can be found deeper within the site.

Testing Service & Full Case Study: Visual Website Optimizer

Reviewing some of these case studies and seeing the A and B versions of landing page elements, you often wonder why the sites weren’t designed that way in the first place?

As we all know (or should know), hindsight is 20/20. Landing page design and A/B testing are a continuous process.

— Angela Stringfellow

default author image
About Angela Stringfellow
Angela Stringfellow has spent the last few years of corporate life working in marketing management in the healthcare industry before transitioning her real-world marketing experience to the web. For the past several years, Angela has worked as a consultant with businesses small and large to build comprehensive social media campaigns, blogging and editorial strategies and enhance overall brand reputation and media presence, with a primary focus on Web 2.0 technologies and content marketing.
» More blog posts by Angela Stringfellow


  1. Rachel Poling

    This is great! I love seeing before & after, plus how much the conversion on each site improved. Thanks for pulling that all together!

    • Oli Gardner

      Glad you liked it Rachel – yeah it really helps to see the thought process and results around an A/B test.

  2. Jens Schriver

    Hey, how cool to see our test featured on Unbounce! :)

    My company ClickLab (based in Brazil) performed the WriteWork test – we are really proud of the results. This particular test was a real break-through for us. It made us realize that knowing your customers and understanding their motivations is fundamental to testing. We probably spent 80% of the project researching and 20% redesigning – and it paid off.

    I’d actually never seen the Obama campaign results before. Amazing test and amazing results!

  3. Srirang (brahmana)

    What is actually interesting is that there are contradicting examples — RJ Metrics had a video and then moved to no-video and some text with bigger CTA item.

    Where as Buy Real Twitter Followers sort of did the opposite thing. They had text (a lot) and then moved to having video.

    It just goes on to say that there is no one size fits all and reinforces the importance of A/B testing.

  4. CK

    Wow this is awesome Oli!
    I am just setting up some pages too…. and I must say you guys developed a great user-friendly system. Congratz!
    The only thing for me is how do I create a more fancy page…. these pages look more web 2.0.
    Anyway keep up the good work!

  5. Jeroen

    Nice examples, the video example in the Twitter case is amazing. We’re all stupid with not working with video…

  6. Anton IMNorth

    Really interesting to see the Obama A/B test. I’ve met with clients that think A/B testing is difficult and can’t produce much difference to the bottom line. So when explaining the real difference it in fact can produce I’ll make sure to show them the Obama statistics.

  7. Bill de Mooy

    I was ‘tweeted’ by Christian @SmartboyDesigns of this posting. My hat is off to you. Well written. It whets my appetite to learn and do more with A-B testing. It’s just that I’m too busy doing my work to do this work.

  8. How to: Calculate the ROI of Conversion Rate Optimization

    […] landing pages. Some companies report 34% + conversions by incorporating testimonials. Unbounce reports conversion lift of up to […]

  9. Ultimate Cheat Sheet: 21 Lead Generation Ideas You Need To Know | ENTREPRENEUR'S WORLD

    […] Obama raised an additional $60 million by simply A/B testing multiple marketing campaigns to see which performed […]

  10. Use A/B Testing to Expand Your Digital Reach » Shaping the Game

    […] A/B testing is a data-driven tool used to determine which version of a web page is most effective in bringing site traffic, and ultimately more opportunities for conversions and revenue. Rather than guessing which digital marketing strategy is more effective, A/B testing is a quantitative method for identifying which modifications will truly optimize an online user experience. A/B testing is commonly used when businesses are considering making changes to their website. The changes can be as small as a different placement of a button, or as large as a complete redesign of their home page. Once the changes are made, half of the business’ website traffic is exposed to the website without the modifications, and the other half is exposed to the website with modifications. Traffic behavior, such as bounce rate, conversion, etc. is analyzed on both versions of the website through a statistical software like Google Analytics or an A/B testing software, and the business can use the results to determine whether or not the website modification had a positive, negative, or neutral effect on site traffic. A surprising yet successful example of employment of A/B testing was performed by Obama’s 2008 Campaign Committee. The Obama Campaign Committee A/B tested two versions of Obama’s campaign website in order to see which version would drive more campaign donations. In version A, the homepage displayed an image of Obama with a “Sign Up” button, and version B displayed an image of the whole Obama family with a “Learn More” button. The modifications were small, but the differences in conversion were massive. Results showed that version B brought a 40% increase in sign-up rates which led to 2.8 million more email addresses collected and an additional $60 million in campaign revenue according to Unbounce. […]

  11. A or B? Which website is best? – imarketdigitally

    […] Check it out here: http://unbounce.com/a-b-testing/5-businesses-that-used-ab-tests-to-lift-conversion-rates-by-up-to-21… […]