Creating an AB button color test is very simple and results can be amazing in some cases. What color a button should have on a website has been one of the long-standing debates in the world of conversion optimization. There are trends in society that are based on color. In fact, apart from the standard red and green that are used to dominate web design, now developers are using colors like light blue, orange, and even pink. We have found in some tests that purple increased conversion by 75%.
However, which one is the best color, is one of the concerns for many developers. The good news is that you can choose the most suitable color by testing different colors on the same page. A test like this was performed on Hubspot, and the results showed that each button color had a different impact when it came to the overall conversion of a page.
Here red and green that are the basic and the most popular button colors were tested. First they created a standard home page with a green colored button. Then they made a clone of that page and the only change in it was the color of that button was made red.
Green Color Button
The green color goes with terms like ‘natural’ and ‘environment’, and with its wide use in traffic lights, it also suggests the idea of ‘Go’.
Red Color Button
Red is traditionally used to communicate excitement, warning, passion, blood, and the stop light at traffic signals. On top if that, it is eye catching. However, it is not as often used as the color green is.
AB Button Color Test Hypothesis
Green means ‘Go’ and Red means ‘Stop’. Do these connotations affect people’s decision to click? The hypothesis made was that even if there was a color that would be clicked more by users, this difference would be very insignificant.
AB Button Color Test Results
Through the test, it was found that in total there were 2000 visits t o the pages. It was also recorded if every visit leads to people clicking the button or not. The results showed that red outperformed green by 21%. That means that more people clicked on red than on green even when the content on the page was the same. This may not be a hundred percent, but even this much was a staggering difference than what was expected.
It is also interesting to see that it is tough to find the same results with user testing. It takes many trials to find statistically valid results. That is why if you are going to keep the samples low, you need to come up with a balanced approach of testing to get appropriate results.
We cannot always generalize such results to every situation. What we can confirm is that they are true for conditions in which they happened. There are tons of reasons that could show you why and how the results came up the way they did.
It is important that you do not go on changing all the green buttons on a page to red; instead, test them out first so that you know what color works on your page for your audience.
Check out this great graph explaining colors:
If you need help in creating an AB button color test, see our product Reactor for more information on how we can help.