A Data Science Central Community
Will a red call to action result in a higher conversion rate than a green one? One of the most controversial topics in web design is the issue of color. This subject attracts a great deal of attention, based on the notion that the color of an object can affect the way we feel about that object.
Color is shown to be a signiﬁcant determinant for both website trust and satisfaction. Color has the potential to communicate meaning to the user and inﬂuence the visitors’ perception through the priming effect.
This is when the exposure to one stimulus then influences the way we response to a further stimulus. In this way the exposure to a certain color can influence the visitor’s reaction towards the site in a ‘carry-over’ effect, meaning that the emotional reaction towards a color can be translated to positive or negative interaction with the website.
There have been numerous (unsuccessful) attempts to classify consumer responses to different individual colors. Those finding reveal that the reaction towards color is more determined by personal experiences. As a result, it can’t be universally translated to specific feelings. What this means is that our reactions to color are not absolute. So there is no collective preference for one color over another. So in our initial example, there is very little evidence to support that a ‘green’ call to action will universally make people purchase a product more often than ‘yellow’ or ‘red’.
DMIX ran an A/B test for CareLogger, a health app that allows users to keep track of their diabetes. They tested call to action buttons in two different colors to determine which attracts more clicks. “After 600 signups with the exact same call to action and just a different color button, the red version saw 34% more conversions than the green”.
After seeing these results, most designers would then conclude that the red CTO leads to a higher conversion rate. But, an analysis of a similar example’s heatmap and session replay showed that that conclusion was based on partial information only.
The true reason for the difference in conversion is often due to the buyer persona rather than the color: Those that convert tend to be more ‘impulsive’ buyers (see the left hand heatmap). The analysis showed that they progressed through the page and the form quickly without attention to details, while those that failed to convert were more focused on the content of the page (the right hand heatmap).
Now that we have this vital information, action items for improving the conversion of the site could include:
Thus color was a questionable part of the overall online customer experience and far less important in contributing to the overall optimization and conversion.
Anthony G. Greenwald and his colleagues observed that people make connections much more quickly between ideas that are already related in their minds. For example; the strong cognitive link between man and soccer. Or women and makeup.
We are primed to connect ‘man’ and ‘soccer’ due to their strong connection in our mind. These connections were developed through the socialization process and are culture related. But we can also use this concept to connect colors with products or offerings.
Our reactions towards colors stem from the conditioned links between a certain color and what this color represents.
Thus, if pink is automatically linked to pretty little girls, this well established connection can be used to communicate a brand’s desired image in the consumer’s mind (Madden et al., 2000). For example, if a new website for girls clothing uses pink, it actually takes advantage of everything that this color stands for.
At the end of the day, there is no set of hard and fast rules that can guide you through the process of choosing the right color for your websites. But, color should be connected to the specific message one wants to convey. Black can fit nicely for a business website but not for a wedding website for example. The usage of a specific color should try to take advantage of our cultural experiences and rely on our existing associations.
This article was written by Liraz Margalit and was originally posted on the ClickTale blog.