In 2013, Wikipedia needed to do some fundraising to help keep the company on its feet. The money raised would ensure that necessary aspects of the company, such as the servers, could continue to function. Those in charge of the fundraising campaign decided to do some split testing utilizing different banner designs to test visitor reactions. Let’s take a look at what they did, and how it affected their income.
Highlighting – Cyan vs Yellow
Image via Wikimedia Commons
Wikipedia started the testing trying to learn if there was a difference in the color used to highlight some of the text in the banner. Each of these banners were yellow, one with a cyan highlight, and the other with a brighter yellow highlight. These banners were tested in July and August, and they declared the yellow highlight the winner.
Cyan Yellow Total Donations: 4,337 4,418 Total Clicks: 7,127 7,226 Maximum Donation Amount: $500 $500 Average Donation Amount: $12.33 $12.53
Display – Form vs No Form
Image via Wikimedia Commons
In May of 2013, Wikipedia decided to try out their banner split testing to find out whether they would get a better response with or without a form. The form made it easy to select how much a visitor wanted to donate, as well as how. They decided to go with the form variant.
Form No Form Total Donations: 30,768 28,145 Total Clicks: 45,970 39,919 Maximum Donation Amount: $759 $1,000 Average Donation Amount: $13.31 $13.61
Color Differences – Gold vs. Grey
Image via Wikimedia Commons
In December of 2013, Wikipedia split tested with two fundraising banners – gold and grey. The text in each banner was the same. The company hoped to raise $20 million between the two banners. The stats below revealed that gold was more effective:
Gold Gray Total Donations: 1,918 1,860 Total Clicks: 3,657 3,694 Maximum Donation Amount: $500 $110 Average Donation Amount: $12.63 $11.35
Outcomes
Image via Flickr by Giulia Forsythe
One of the most important lessons we can learn from this case is that split testing is important. Wikipedia decided that it was a good idea to experiment with different wording, colors, and types of banners throughout the year. The information they learned allows them to narrow down options to find the best banner for their fundraising projects. As they look through the information presented by the stats, they will be able to see that the gold banner did better overall, form will get more clicks and donations, and yellow highlighting did better than cyan over all. They decided to roll out that final variant to a majority of their traffic during the peak campaign time.
No matter what type of website you have, or whether you’re doing a fundraising project or selling a product, split-testing is a viable option for helping you understand what your customers like to see, and what is most effective for your product, service, or request.
Contact us today to increase your Conversion Rate . . .
We put our money where our mouth is – click here to contact us today and learn how we can drastically increase your website conversion rate for no money up front, and no fees until we get you an increase!
By Jon Correll