The Presentation Style of Web Analytics Data and Decision-Making is a condensed version of my full dissertation that looked at how the style in which web analytics data was reported had implications for the ability of digital marketing professionals to confidently make effective business decisions.
About the research
The research focused on a particular reporting style, commonly labelled as an infographic but named by some as an "infoposter" and examined its validity as an effective reporting style for web analytics as well as its effect on the ability of digital marketing professionals to make decisions.
The report also suggested a potential definition of the style for use by future researchers, based on related research into the effect of data visualisation on decision-making.
It was my belief that an infographic ("infoposter") would have a detrimental effect on decision-making ability and quality.
The validity of this hypothesis was tested by conducting an experiment on two groups of digital marketing professionals: comparing their ability to make a decision and the quality of their decision-making process, when given a web analytics report presented as either raw data in a tabular format or as an infographic ("infoposter").
The key findings
The research implied that there was no discernible difference in the ability of digital marketing professionals to make a decision using either data in a tabular format, or as an infographic ("infoposter").
There were subtle differences in the quality of the decision, but the influences on this could not be confidently determined, potentially due to limitations imposed by the final sample size.
Why infographics ("infoposters") might not help decision making
Data visualisation at its basic level should clarify the underlying data, helping to communicate it in a way that supports the decision-making process. The research showed that the "infoposter" style differed from a "true" infographic in many ways, not least the use of graphics.
Typically "infoposters", as Connie Malamed observed, "collect facts and figures about a topic" and were designed to tell a story or communicate an idea or opinion in a highly graphical fashion. They tended to rely heavily on text and numbers to inform rather than where absolutely necessary and place emphasis on their design over their ability to communicate the underlying data clearly.
The addition of graphics as pure decoration (Edward Tufte's "chartjunk") detracted even more from their use as an effective communication tool. Ziemkiewicz and Kosara observed in 2007 that the most effective data visualisations had "bijectivity" where "every piece of information corresponds to exactly one unique visual element" and this was a "necessary condition for a visual representation of information to be readable".
It was also shown that infographics/infoposters were not preferred by digital marketing professionals when receiving web analytics reports. The dashboard style was by far the most preferred by the sample group used in the research, as shown in the chart below.
What are the implications for both web analytics data and for infographics ("infoposters")?
The research also suggested that the range of metrics used by digital marketing professionals to make decisions was quite narrow, reinforcing the findings of the recent Econsultancy and Lynchpin Online Measurement and Strategy Report. This report found that the 59% of the companies surveyed used 50% or less of the web analytics data collected for driving decision-making, as illustrated in the chart below.
My opinion of "infoposters" has not been changed by this research, I still see them being of little use when reporting web analytics data, but do believe that (when done well) they have their uses as a marketing tool.
What are your thoughts?
Do you feel infographics ("infoposters") can help in web analytics decision making? Have you ever used such a technique to report insights to senior management? Or do you think that infographics ("infoposters") will never rise above their status as linkbait? Share your views in the comments below.
Future blog posts will look at elements of this research in more detail. For more information or to ask any questions about the content of this report or the full dissertation, please contact .
No hay comentarios:
Publicar un comentario