Day 2 is upon us and we presented our dashboards this morning. As of now, we are on our second day and its dealing with APRA Superannuation data. APRA, the Australian Prudential Regulation Authority, is the regulatory body that oversee the superannuation industry in Australia. They have collected data over the years on various metrics like: financial performance, demographic breakdown by age and gender, and many others.

Today, our time has been shortened due to having presentations in the morning. Not only that, it being superannuation data a lot of the terminology was very illusive. As a result, I didn’t want to waste anymore time in terms of trying to join data and kept it as simple as possible.

Learnings from Day 1

I didn’t particularly time-box well. My plan for Day-1 was to finish by at least 6PM however, I ended up going home and didn’t finish everything I wanted until around 9PM. The major theme of dashboard week is time boxing, and for Day-2 I would like to do this a lot better. Making things more difficult the time lost in the morning means that de-scoping is a priority.

Data Prep

I decided to stick with just cleaning one table. Rather than trying to everything at once I decided it would be best to stick with one and try to achieve some insights from it. I chose Table 12 from the data, and it being an excel file was quite messy, and needed quite a lot of prep.

Above is Part 1, where I’m just getting the data and dropping columns that are not necessary.

Above is Part 2. Due to the mixture of dollar value and count figures of members, I had to separate them into its own table. Which I joined towards the end in Part 3 (see below).

After, joining them a hyper extract was ready to be visualised in Tableau.

In terms of insights, Table 12 (the data that I’m looking at) is centered around demographics with respect to age and sex. I’m thinking about if there is any disparity between the sexes in terms of average super balance, as well as if any particular age groups tends to have more than others.


My main goal was to see if there were particular industry types that certain age groups would move to. For example, do people <25 years go from an industry to a retail super fund. My findings are that you can see a shift from Industry to retail superfund as people get older. This could be due to various reasons like retail may provide better returns versus industry, as the realise they need more money for their retirement.

A snapshot of the dashboard:

In order to improve, I would consider looking at more tables like fees and investment performance to do a more in depth analysis.

The Data School
Author: The Data School