In this Amazon QuickSight introduction, we will show you how to build a dashboard and make basic data visualizations. Amazon QuickSight is a cloud-based data visualization tool provided as a service with pay-per-session pricing,
We were curious to see what it would take to create a QuickSight visualization similar to the visualization created in our earlier blog post, “Embedding Power BI Reports in a WordPress Blog”. The visualizations show the TOP 5 registered female and male first names per state, as per SSA.GOV for 1910 – 2018. The entire dataset is available as a zip file. The zip file contains comma delimited files for each year since 1879 as of current date.
The entire process of loading the data into AWS, preparing the dataset for QuickSight consumption, and creating the visualization was pretty straightforward, albeit a few challenges, which I will get into later.
Creating a basic data visualization with Amazon QuickSight
Here is the process, I followed to build my first data visualization with QuickSight.
To begin with, I created an AWS Free Tier account and downloaded the zip file of data which I unzipped into a directory. I created an Amazon S3 dataset and loaded the data files into this dataset.
After that, I created an Amazon QuickSight account in the US West (Oregon) region and purchased 1GB of SPICE Capacity at a cost of $0.25/GB per month.
Next, I created a JSON manifest file, where I specified which files from the Amazon S3 we need to import into Amazon QuickSight. I uploaded this JSON manifest file into the S3 bucket containing the data files.
After the data files were specified, I set up a QuickSight S3 data source pointing to the JSON manifest file.
With my next step, I created an analysis within Amazon QuickSight. In order to filter to the TOP 5 male and female names, I created a calculated field called Frequency Rank that ranked the names by descending frequency. You can find more info on QuickSight calculated fields here.
For users could change the TOP n value, I added a parameter with a control to the value parameter, which means that this parameter is “controllable” from within the visualization. And the third parameter I added was Gender, which we need to filter the Top n tables by Gender.
This is how the applied filters for the Top 5 Male Names table looks like:
And here is the report I got after all.
Observations on Amazon QuickSight
This was my first QuickSight report. I found the tool fairly easy to get started with, but there were some challenges. The idea of parameters having controls is an improvement over earlier versions.
Next, I will incorporate additional datasets into the visualizations to see how QuickSight performs as a data modeling tool. Stay tuned!
