This week is dashboard week for DS22.  What does it means? Simply put, every day is project day. Which is the by far the toughest day of the week. Sounds fun, am I right?

Let's take a deeper look at what this dashboard week is.

We are assigned a new project every morning, we might be given a data set or we might have to look for the data to use. Then we have to prep the data and create a dashboard. Each of us works individually but we can ask for help on Convo. In the meantime, we have to write a blog post, and possibly, learn something new along the way.

Today we have been given data about the Greenhouse Gas Emissions around the world. I really wanted to make an informative dashboard with accurate data and simple to understand, that is why I took extra care in the data prep process. I spent some time reading the documentation available on the website where the data lied (https://oasishub.co/dataset/global-cait-emission-projections) and I also did a bit of research across external sources to be sure I could understand all the information provided. Then I opened the data on Alteryx and started exploring it to think about what I wanted to present and which information I needed.
The data was already in a pretty good shape, there were not much preparation to do, a part a bit of cleaning and selecting the fields for the purpose. I did a couple of tries with the workflow but in the end I decided to focus on EU27 countries. With a bit of extra time I could have had homogenised certain inconsistencies across data that existed due to the different recording systems used across the year (the file contains information collected across 167 years ).  Pivoting the table allowed me to output the data in the shape I need for the analysis.

At the end of the prep I went from 21945 rows and 174 columns to 1352 rows and 4 columns. Now I am ready to jump on Tableau.

Since we are asked to do a KPI/informative dashboard, I thought that showing the emissions trend over time would work well with a line chart, and I also created a series of spark lines to split the view into its component. They are also paired with bar charts which sizes represents the percentage of the total. I created a BANs sheet to collect the most important information (and I successfully used a LOD expression - wohoo!). For the BANs in the main line charts I chose to play with labels and annotations. Here's the result of this morning of work:

I am happy with the overall look of the dashboard, but I think I still have to work on my time management skills - I took some extra time to look at the data, which is surely good practice when some might not be sure about the data, and I really didn't want to present incorrect information, but I had to rush through the dashboard and I am afraid that gives me little time to experiment. But it is just the first day of the week, so I hope I can be better on this by Friday!

Update: after implementing Andy and Carl's feedback, this is the dashboard.