The August monthly challenger over at Storytelling with Data was one I learnt a lot from. The challenge was to find a publicly available dashboard, analyse it, find the story and then build a graph that summerizes this story.
My dashboard experience
This was particularly intresting to me because in my work I build dashboards a lot to show how a website is performing against business objectives. And these dashboards aren’t meant to provide a deep analysis, they are a quick overview of “what is going on”. They are the beginning of the story of the available data. These dashboards are usually a snapshot of website performance against conversions and if we see website conversions decline over time then we know we need to dive into a deep analysis.
I build a lot of so called real time dashboards. These dashboards offer an up-to-date view of the website performance and can be adjusted to a different date range. This will allow me to identify trends and start my story of why a certain trend is increasing or decling.
In my experience that is how I would define a deashboard – a tool to give you snapshot of overall performance that will then allow for a deeper anlaysis.
What is a dashboard
One of my favourite books on data visualization is Information Dashboard Design by Stephen Few. His book is a great resource for anyone willing to master the art of building dashboards. Working on a dashboard is a lot more different than building a single chart or a presentation. In one sheet you need to give your audience an overview of performance and catch their attention. Here is how Stephen defines dashboards:
A dashboard is a visual display of the most important information needed to achieve one or more objectives, consolidated and arranged on a single screen so the information can be monitored at a glance.
Using a dashboard to find a data story
Working on this challenge on the Storytelling with Data community helped me with critical and analytical thinking. These are very important skills to have in any analytical job in order to tell data stories. I found this Tablaeu dashboard by Nick Adams which shows the median annual salary for registered nurses by state and how this has changed from twenty years ago, as compared to 2001. The dashboard is very well designed and executed. It is interactive and allows people to select a state from the list and view detais for each year since 2001. The dashboard also summarizes the total absolute increase from 2001 for each state.
I wanted to put all this into a single chart and highlight the story I am taking from this dashboard – and that is which state pays nurses the highest and which state pays nurses the lowest. I also wanted to find a way to visualize the increase from 2001 to 2021.
I built a slopegraph in Flourish. Since I am plotting a large set of data the chart still needs to offer some way for interactivity. The slopegraph plots all states, what the median salary for registered nurses was in 2001 and what it is in 2021. I am highligting the top two states on the graph, California and Hawaii, and the bottom state, Alabama. California and Hawaii pay registered nurses the most and have also increased the median salary the most since 2001. The interactive chart in Flourish offers hover over effects to highlight other states and also choose a state from the drop down menu. This can show the audience where a certain state sits between the highest paying states and the lowest paying states.
As with any visual design the fundmental goal of a dashboard is to communicate. And to communicate effectively to your audience dashboards should be easy to understand and they should be understood clearly and quickly. Whatever the data story is in your dashboard people should be able to easily find it, decipher it quickly and act on it with accurate information and insights.