What is our primary use case?
We build nonprofit organization analytics using Apache Superset in this system.
I am working only with an on-premises setup of Apache Superset, having set it up locally in Docker Compose. I don't use the cloud version because all data is stored locally.
What is most valuable?
I trust Apache, which is why security is critical for us, making Apache Superset a trustable open-source product. It's easy for clients and for people who haven't used BI tools; it's good for them to start with Apache Superset.
It's very easy, reliable, and trustable because it's an Apache open-source product. In general, I find Apache Superset easy to customize; one case is that I needed to add the ClickHouse plugin. Apache Superset is very useful because you can install everything you need using Python and different libraries, making it very useful in this area.
I appreciate the capability in Apache Superset that allows you to separate queries and dashboards and reuse them.
Apache Superset is a very easy tool to set up and start working with, so I find its UI/UX very useful, enabling me to get started quickly.
What needs improvement?
Apache Superset has good features for supporting multidimensional analysis, but I need more ability to customize. This depends on different tools used for different customers or users; Superset is good for users with limited experience as for technical people and analysts.
It's a good tool, but sometimes you need more customization. Building custom graphs was hard compared to Grafana, which is much easier. Generally speaking, it's a different level of customization, so if you need super customization, you might consider using other tools. But for starting out, it's a good tool.
One thing I would add is folders because having 1,000 queries makes it horrible to understand which query was written and for which dashboard, with graphs. I would want to see more features such as filters and folders, and improvements in how to organize queries and dashboards overall.
I had issues when trying to export and import dashboards; I could not do it on the first attempt. I spent a whole day trying to fix it because I built the dashboard on one laptop, exported it from the web version, and then struggled to import it on another laptop. I believe this process is critical and fixable, but in my case, I could not do it. This led us to consider other tools for continuing our work. It would be great to improve the reliability of the import and export features.
For our tasks, Apache Superset's integration with authentication protocols and role-based access control was sufficient, but it would be great if we could create teams and organizations. It actually depends on the size of the company; for small companies, I think it's more than enough. For enterprise or large companies, I would say it may need more advanced roles and organization features.
For how long have I used the solution?
I have been working with Apache Superset for around one year, a little bit less than one year.
What do I think about the stability of the solution?
In general, I have not experienced stability issues or crashes with Apache Superset, except for the import/export concerns which are critical. Everything else is good.
What do I think about the scalability of the solution?
I can't provide feedback on the scalability of Apache Superset since we don't have a lot of data. Comparing it with Grafana, I am not ready to comment on performance under load because I haven't tested it with large datasets. But based on basic data, I can say that it uses about half the memory compared to Metabase.
How was the initial setup?
The implementation process for Apache Superset was very easy and straightforward.
Which other solutions did I evaluate?
I am looking into Grafana and Metabase as alternatives to Apache Superset. I appreciate Metabase for its tab feature, which helps organize graphs better.
Metabase has some AI capabilities. However, if you're building something enterprise-level, I think AI might not help you a lot; it could assist a little with simple queries but not with complex custom queries, making me unsure about its utility.
What other advice do I have?
I am more using SQL for data querying within Apache Superset.
I read a very good book that gave me a simple piece of advice for organizations considering Apache Superset: it is essential to understand who the users will be for your tool.
When you start working with any tools such as Apache Superset, you should first understand the features and the goals and challenges you will face. You need to know who will be using it and interacting with your graphs.
If you have experienced users with a lot of data, you may consider other tools. But if you want a quick start for your analytics platform or to build something fast, Apache Superset is a good solution.
My rating for Apache Superset as a solution depends on the company size. For small companies, if they fix the import/export issues, I would rate it a 10. For large companies, it may not be suitable, and for medium-sized companies, it depends on their experience with analytics. In general, I rate Apache Superset an 8 out of 10, providing a 10 for small companies and an 8 for large ones.
Which deployment model are you using for this solution?
On-premises
If public cloud, private cloud, or hybrid cloud, which cloud provider do you use?
Other