This week we were taking a trip back to the Tableau Ecosystem, learning about how different softwares and tools work, as well as learning more tools in Prep and Desktop. This week was our final Friday Client Project, before the Week Long Client Projects begin.
Beginning with Tableau Prep, we were taught about the Rank function and it's ability to adjust the type of rank used when data is processed (Find my article on it here). This was followed by LODs (Level of Detail), which work in a similar way to aggregating in that you can calculate sums, averages or maximums across a whole table or specific parts of data. Or in simpler terms at different levels of granularity than is present in the current visualisation. We focused on Fixed LODs, which compute values using specific dimensions. For example, to compute average sale price per sub-product, you would Group by Sub product and compute using Average. After learning to plan our data preparation and the concept of scaffolding, we familiarised ourselves with Tableau Server, and created our personal as well as cohort folder on the server.
The next section we moved on to sets, which help to answer questions about specific subsets of data. These are custom fields defining a subset based on set conditions, letting user interact with a dashboard to control their analysis. We used general, top, combined and conditional sets to analyse data, and were than tasked with recreating 2018s Workout Wednesday Week 43, and Week 38. Afterwards we looked at using parameters for user adjustable reference lines, table calculation, measure and dimension switches. Another exciting technique we learnt was DZV or Dynamic Zone Visibility, which allows you to control which charts show based on a selected parameter.
Final content for the week was an introduction to using Table Calculations, a transformation applied to values selected in a visualisation. They are view based; a special kind of calculated field that computes based on what is currently in the visualisation, ignoring dimensions or measures that are filtered out. We covered the Quick Table Calculations and how to adjust their settings depending on the value we wanted to calculate, running total graphs, first & previous periods, Year-on-Year KPIs, moving average and Bump charts. Whilst a difficult concept to grasp, it is already apparent that this powerful tool will prove valuable for future analysis.
The final Friday Client Project proved interesting, where we were asked in a similar way to the previous week, to create a final concept sketch based on a fellow cohort members plans. But this week, the visualisation took place in Tableau Desktop. This week I was tasked with creating a visualisation based on Car Insurance Claims Data, which proved a challenge forgetting how some simpler tasks in PowerBI are not so simple in Tableau. However, the presentation went well and capped off a good few weeks of learning about the client project process.
The next week kicks off our first Client Project, with the internal project coming first. It will undoubtedly be a testing time, but I look forward to working with the rest of DS53 to create a successful project, and embark on many busy weeks coming up.