Jon's DS13 Week 3 Recap

by Jonathan Allenby

Week 3 covered a real spread of Tableau products including Desktop, Prep, and our first real glimpse into Tableau Server! It was also filled with other exciting (scary) news such as finding out the date for our first teaching-the-public event, as well as finding out our first client project!

Day 1

Day 1 was jumping back into Tableau Prep and learning more about some of the neat little facts & functionalities that won’t jump out to a first-time user. These included:

  • Shadow Extracts: Tableau Prep creates & hides on your computer the extracts for the last 5 datasets you’ve connected to. This is in order to help speed things up if you decide to reconnect to them in the near future. Good luck finding where it saves these.
  • Wildcard Unions: If you have a bunch of similarly named files (e.g. MyData_01, MyData_02, MyData_03, etc) that you want to load into Prep and work with then you don’t have to go to the trouble of loading each in separately and slowly unioning them together. Instead you can use a wildcard union (e.g. MyData_*) to load and union them all together in one step!
  • Group & Replace: Spoiler alert for Preppin’ Data W2, but Tableau Prep has some neat little options that can help catch spelling or text formatting errors that can automatically group values on a variety of factors such as spelling and pronunciation!

Day 2

On day 2 we headed back into Desktop to continue on with Tableau intermediate. We covered some more advanced features such as: sets & combined sets; parameters & quick parameters; and LODs which are Level-of-Detail calculations. LODs were particularly mind boggling and something we’ll definitely need to practice and revise. A very brief example use-case for an LOD is comparing the Sales for each Region against the average sales across all regions. More writing on this to come later!

Day 3

Day 3 introduced us to what I suspect is the most foreign product to us all: Tableau Server. This was our first outing and whilst the first few slides had our eyes boggling a bit, it actually starting making sense and coming pretty naturally – especially after we got hands on with the product. I think the best way to think of Tableau Server for a complete novice is as more structured, private version of Tableau Public. It allows businesses to manage who can view what resources and what permissions they have, share data sources and visualisations, and create a structure for projects or clients.

Day 4

Back at it again with the Tableau Desktop, however this day we officially entered Advanced territory. Today we waded back into the world of Table Calculations and picked up some cool tricks such as how you can figure out how to build certain Table Calcs by making a reference line first, as well as the useful functions of FIRST(), LAST(), and INDEX(). Once again it was also sprinkled with incredibly general advice, such as using colour to highlight positives and improve how information can be received by viewers. We also took a little look into how to represent (and avoid misrepresenting) and read basic distributions of data, and we finished the day with a fun little session of trying to make improvements on a basic barbell chart.

Day 5

For this week’s project we were each given the name of a publicly traded company and tasked with producing a visualisation using their historical stock value data. The caveat for this task was we had to use at least two table calculations, and the visualisation should be designed to convince the reader to buy stocks in our company. Within the 3 hours we had my fellow cohortians produced some really fantastic visualisations and you could definitely see the progress we’ve all made since the first week. I myself had a bit of a stumble on this week’s project but I got more out the experience than from anything else so far. The Data School is not only designed to help you reach and exceed your maximum potential, but also as a safe environment for you to struggle, worry, and fail. This is something you don’t want to experience for the first time with a client and if you don’t experience failure it can be difficult to learn how to pick yourself back up. The support on offer for me this day was immaculate, and I hope the takeaway message for this is that it’s okay to not feel okay and help is always on hand at the Data School.