Original


Source: The 21st Century War of Online Ads, 2020.


Objective

The objective of the original data visualization is to indicate an increase in the revenue generated by online advertisements over the years and also highlights the growth and decline of the percentage revenue earned through different advertisement sources. The target audience is the general public specifically those internet users who are unaware of the fact that they are unknowingly contributing to the growth of the revenue earned through online advertisements.

The three main issues identified in the visualization are as follows:

  • The information presented in the visualization is misleading to the audience as there is no flat baseline for comparing the values in the graph for different categories. The visualization uses the height of each coloured stack area to represent the percentage proportion of the online advertisement category at a given point of time. The inclination to interpret the top of the area as a quantity makes it difficult for the audience to interpret the trend for the different categories as the viewers may be drawn to the changes in the slopes of the coloured area creating an optical illusion. Also it is quite improbable to perceive the values at each time stamp. In the graph, the percentage of mobile ads revenue appears to have increased from 40 to 65% (approx.) however in reality the actual revenue percentage rate increased only from 5 to 25% over the period.
  • Extensive use of colors to represent the different online advertising types makes the visualization appear quite messy and poses too much of information for the user to interpret the basic structure of the visualization. Although the visualization provides a label for each type of advertisement , it is difficult to discern between the colored stack areas of the referrals, email and sponsorship advertisement category thus contradicting the principle of creating accessibility through intuitive design.
  • The visualization fails to provide an accurate representation of the actual data as the percentage total has been computed incorrectly for the years 2008 & 2009 as the total percentage shown calculated was 99% & 101% respectively. This information does not reflect in the visualization and hence fails to follow the principles of ethics.

Reference

Code

The following code was used to fix the issues identified in the original.

library(ggplot2)
library(readxl)
library(tidyr)
library(dplyr)
library(magrittr)
library(colourpicker)

ads <- read_excel("ads.xlsx",sheet="values")

ads1 <- ads %>% select(year,keyword_search,mobile,display_ads,digital_video,classifieds,lead_generation,rich_media,sponsorship,email,referral,others) %>% gather(key = "ads_type", value = "value", -year)

ads_names <- list(
  'classifieds'="Classifieds",
  'digital_video'="Digital Video",
  'display_ads'="Display Ads",
  'email'="Email",
  'keyword_search'="Keyword Search",
  'lead_generation'="Lead Generation",
  'mobile'="Mobile",
  'others'="Others",
  'referral'="Referrals",
  'rich_media'="Rich Media",
  'sponsorship'="Sponsorship"
)

ads_labeller <- function(variable,value){
  return(ads_names[value])
}

plot1 <- ggplot(ads, aes(x = year, y = Total)) + geom_line()+geom_point(size=1)+scale_y_continuous(limits = c(0,100))+labs(title="Rise in the online advertisement revenue ",subtitle = "(Total online advertising revenue generated over the years 2000-2014)",x="Year", y = "Total ads revenue (in billion dollars)")+theme_bw()

plot2 <- ggplot(ads1, aes(x = year, y = value)) + geom_line(color="#325B84",size=0.5)+geom_point(color="#325B84",size=0.2)+scale_y_continuous(limits = c(0,20))+
       labs(title="Revenue generated using various methods over the years 2000-2014",x="Year", y = "Ads revenue (in billion dollars)",caption = "Source: The 21st Century War of Online Ads, 2020")+
       facet_wrap(~ads_type,scale="free_y",labeller=ads_labeller)+theme_bw()

Data Reference

Reconstruction

The following plots fixes the main issues in the original.