Case Study: 250,000 Customer Split Test Leaves No Doubt
Jan 2nd, 2009 by Scott Brandley
Trust Guard Improves Conversions by 16.51% with Just One Seal!
When your running a split test, the larger the sample size, the more accurate the results. It also happens that more traffic gives you the ability to test more variables, as was the case with our latest split test at aplaceformom.com. Aplaceformom.com is one of the largest online lead generation websites for retirement homes across the country, getting about 8,300 visitors a day, or 250,000 visitors a month. Because of their high traffic volume, we were able to do something unique – we split tested 32 Trust Guard seal combinations all at the same time to find the highest converting seal for their particular needs.
While most of our clients use multiple Trust Guard seals on their site to maximize conversion, which is the ideal configuration, aplaceformom.com is a lead generation site and not a store, so they only qualified for two seals – the Privacy Verified seal and the Business Verified seal. In addition, they only wanted to display one seal, so we created 32 different seal options in our test – 16 Privacy and 16 Business – all with different colors, sizes, and styles – and then we rotated them in a single position on their site, directly underneath their application form.
In addition to the 32 seals, we also rotated a blank option, where no seals showed up. In the first round we narrowed the test down to the top 8 performing seals (9 including the blank option). In the second round, a clear winner began to emerge to the point where it beat all other options, but most importantly, it beat the blank ‘no seal’ option by a whopping 16.51% with a 98% confidence interval. The winner was the small blue Business Verified seal. By adding this seal to the bottom of their application, they now receive several thousand more applications per month than before. Not too shabby.
So, should you go and buy the small blue Business Verified seal for your site? Not necessarily. While this case study shows that a single seal can dramatically increase your sales, having multiple seals is always more effective because you’re able to resolve multiple customer concerns instead of just one. It’s also important to remember that each website is unique and has different demographics, so what works on one site may not work on another, which is why we have different colors, sizes, and styles available for you to choose from.
Looking back, this was a pretty fun test. I want to thank Ben Gilbert at aplaceformom.com for his willingness to participate in our test. It’s always great to work with people like Ben that can see the incredible value in split testing. The key to split testing is to keep doing it. Not every test is a winner, but by continually testing, you find winners, and you permanently improve your conversion every time a winning test is completed. I also want to thank our programmers here at Trust Guard for their ‘mad programming skills’ in putting together a test of this size and scale. For all of our future clients reading this, on behalf of myself and the Trust Guard staff, we look forward to helping you increase your conversion.
