It’s day one of Dashboard week for DSDU3, and our opening challenge is to use the Domain API to download Australian housing market data and build a visualisation.  Having recently relocated to Sydney from the UK, I have recently and spent many hours trawling through Domain listings trying to find the perfect flat. As you can imagine I was very excited to see what tools I could build to make any future relocations easier! I also have friends moving over from the UK in September who are working in the Eastern Beaches suburb, so I decided to use the Domain API to generate rental property listings for this area.

Generating API’s Tokens and URL’s in Alteryx

Most of the morning was spent reading Domain’s API documentation and configuring workflows to generate the required URL’s and header information for the Alteryx download tool.

Once a Bearer token was generated using the download tool, I built a post request that enabled for 200 listings to be generated per post.

I then used the union tool to combine the resulting listings together before JSON parsing and transforming the data in a lengthy Alteryx workflow.

Once I had a complete data set, I followed the standard JSON parse, field selection and crosstab process to transform my data into a useful structure. I then parsed various fields of unnecessary characters and punctuation to give the desired numeric field output.

In addition, I downloaded the cover picture for each listing to my Tableau Custom Shapes Repository. I configured the download tool to name each picture with the unique listing ID. This allowed me to assign the pallet directly to the listing ID without individually mapping each image to the record, and for the inclusion of the picture in tooltip for listing.

Visualizing the Data in Tableau

For my visualisation, I wanted to try and identify good value rental properties within the Eastern Beaches region of Sydney. This is a notoriously expensive area to rent properties, due to the ease of commute to the CBD, beautiful beaches and generally high standard of property.

My dashboard allows for the filtering of listings by postcode, suburb, the number of bedrooms, bathrooms, car spaces and most importantly, price!

I used a LOD calculation to determine the average listing price by suburb and number of bedrooms and then compared listings against these benchmarks to help identify any potential bargains in the area.

Link to Tableau Public Viz:!/vizhome/SydneysEasternBeachesRentalPropertiesExploration/Dashboard1

Any feedback on my Dashboard is welcome!

Thoughts After Dashboard Day 1

  • Spend less time in Alteryx and more time Dashboarding – 3 pm is too late to start!
  • API’s are tricky to configure but very rewarding when they work!
  • Dashboard week is great and I’m excited to explore different data sources each day!
The Data School
Author: The Data School