The fusionSpan Blog

Create Better Membership Dashboards With Apache Superset

By Alan Luo |March 11, 2021
Membership

Dashboards have always been a crucial part of analytics. As the demand for Business Intelligence (BI) tools increase each year, the amount associations spend on BI products has followed suit. Robust platforms, such as Einstein Analytics and Domo, mainly charge fees by user count. This in turn might limit the number of users that can see visualizations. Should an association have hundreds or even thousands of employees that need to look at internal data, being limited by budget towards BI tools is not a great way to start your data deep dive.

Why Use Apache Superset

We Analysts here at fusionSpan have built our fair share of dashboards. When comparing the different BI tools we have used in the past, Apache Superset would often come out on top due to its flexibility in allowing mass distribution of data visuals without extra cost.

In this article, we have built a membership dashboard to showcase the functionalities of Superset 1.0, a new version of Superset that has just been released this year. We showcase key features of Superset that are used from start to finish, so that readers have a clear understanding of what the tool can do. Now let us jump in!

New call-to-action

Logging In With SSO

Superset supports oauth2 out of the box. This means that web developers can integrate SSO along with Superset. In the screenshot below, we have incorporated Google SSO into Superset. Users can register or login using their google account credentials. This helps increase security drastically, as malicious hackers without emails ending in their .org domain would not be able to login and see any data.

Intuitive Navigation Page

Navigation is important, as it allows users to quickly find the data visualization they are interested in. Superset’s Main Page displays a list of most recently accessed, created, and edited resources such as dashboards, datasets, and charts. From this page, users can also quickly navigate to charts and start building whichever visuals they would like to see.

Upload CSV/Excel Files

Superset, by default, connects to databases and data warehouses. However, if a user had an excel or csv file that they would like to analyze, they can upload that file into Superset and use it to create a dataset. They can then build a dashboard on top of that dataset, or even share that dataset with other users.

Exploring Data in Charts

Superset allows users to easily add calculations and conditions to their dataset, and then analyze that data in charts. In the example below, I am counting how many people have signed up for 2021 membership by looking at the membership field, and then filtering my dataset by the ‘member’ string. As you can see, by using a few clicks, I am able to calculate how many of my Contacts are 2021 members.

Building Dashboards

After creating a few charts, I can now copy and paste these onto my dashboard. As an opensource solution, Superset might not be visually appealing compared to commercial competition. However, functionality wise, this dashboard can provide deep insights to help viewers create actionable business plans.

To explain what is happening in this dashboard, this dashboard displays all contacts that has 2021 membership. I added a filter on top of this dashboard filtering 2020 memberships. This means that the number ‘222’ under “Total Member Count 2021” is now showing the number of people who have signed up for both 2020 and 2021 membership.

Other panels within the chart are showing the average annual spending, as well as the average ages and communication preferences of these members. If we would like to identify who these people are, we can also display a raw data view of their information in a table. The filter will still apply in this case.

Embedding Dashboards Within Websites

Dashboards can be easily embedded in a website or shared via an email or URL link. Embedding dashboards involves creating iframes, which are simple to achieve. Security wise, if the dashboard’s privacy is not set to public, only selected users would be able to see the contents within the dashboard. This means you would have to log in to see the data.

Quickly Combine Datasets using SQL Lab

As I have mentioned in a previous Superset blog, SQL Lab can handle sophisticated data transformations. Using SQL lab, superset users can quickly combine different datasets, allowing users to join and create new datasets using simple SQL queries. In the example below, I have combined Contacts and HubSpot clicks to allow me to see how many clicks my contacts have made in the past.

What is Missing in 1.0

While Superset 1.0 is much more visually appealing and functionally superior to its previous versions, we must be honest when it comes to the shortcomings of Superset. One thing we noticed that Apache has done is remove chart metadata. This is a step backwards, since adjusting metadata allows us to change the color of the chart legends freely and lock filters (so that certain numbers on the dashboards will not be affected by the filter box).

In addition to that, despite SQL Lab being plenty useful in data transformation, we are not able to conduct cross database queries. This is understandable, since Superset itself is not an ETL tool and cannot join queries between two different data sources. To get around this, we have had to combine Superset with other ETL tools such as Stitchdata or Tibco Cloud Integration. Current commercial solutions such as PowerBI, Domo and Einstein Analytics all include their own built in ETL tool.

In Conclusion

In the current day and age where dashboards are being priced up, Superset 1.0 offers an open-source and cost-efficient to build dashboards. It might not contain many wowing factors in data visuals, but it’s flexibility in sharing and displaying data gives it a powerful edge over its commercial competitors. We look forward to the future versions of Superset, and will continue to use them for data analytics.

For your dashboard and data inquiries, don’t hesitate to reach out to our talented team of Data Analysts here at fusionSpan today!

New call-to-action
Alan Luo

Alan has been working in the field of Ecommerce for 4 years and has extensive knowledge in Online Marketing and Data Analytics. He has practical experience scraping and cleaning data and has tackled a few data science projects in his career. During his free time, he would listen to 80s pop and J-rock, or would try a few riffs on his guitar.

More posts