Custom polygons in Tableau

The Department of Veteran Affairs was seeking a more effective way to realise key demographics visually over specific spatial areas like post codes and electorate areas. In charting terms, two metric variables were age and client numbers and categorical variables of location, state and the problematic postcode and electorate boundaries. The challenge was to realise electorate and postcode boundaries that are not native areas for charting applications like continents and states. As it turns out Tableau handles spatial data formats quite well using custom polygons…

Firstly, map info files in ASGS format (the spatial standard used by the ABS since 2011) were fed into a third-party software such as Alteryx that basically (long story short) transforms the shape file into something that Tableau can read. The main consideration here is the creation of vertices (points) and centroids (centre points) coordinates used to map spatial data using polygons. Basically, this information gives Tableau the necessary information to create a dot-to-dot shape for a map overlay based on the vertices in relation to latitude and longitude coordinates.

Next, the spatial dataset was merged with the metric dataset that enabled us to establish the relationship between client demographics: metric and categorical data sets.

For the chart below, a dual-axis chart with transparency was created to show the relationship between client age and client numbers as a heat map per electoral location – The warmer areas are those with high client age and quantity. The scatterplot chart (bottom) allows users to spot the outliers and drill further into the data.

Note: the charts are interactive, you can tweak them and drill-down into the data.

Whereas, this version shows users the Department of Veteran Affairs office locations and the rate-of-change chart (bottom) over how client metrics are tracking over time.

Indeed, the same methodology can be applied to any of the ABS spatial-structured datasets such as, capital and regional cities, indigenous, urban rural areas as well as administrative regions such as electoral and postal areas. Tableau also handles a host of other file types listed here.

The addition of spatial analysis for metric data can give your organisation an edge in realising key demographic intelligence factors, to drive operational performance and optimise expenditure. Find out more by contacting a Buildings Evolved consultant here.

Share your thoughts