1. Every Monday - Same Old Story (a story of our full-stack dev)

You come into work and you're already drowning in messages:

1. "Can you pull the data from last week?"
2. "Can you break it down by region?"
3. "Can you compare it to last month - quickly, we've got a meeting in 20 minutes?"

I'm a developer. I was hired to write code, not to be a living embodiment of Excel. But Slack doesn't care what you were hired for.

The solution came from an unexpected place. I stumbled across Apache Superset - and since then, it's been almost quiet.

2. What Is Apache Superset?

Apache Superset is an open-source BI platform originally built at Airbnb and now maintained under the Apache Software Foundation. In short, it's a free tool for data visualization, dashboard building, and SQL querying - all in the browser, with nothing to install on your colleagues' machines.

I'll only briefly compare Superset to paid tools, since a direct feature-by-feature comparison isn't really fair given the difference in scope and target audience.

The workflow is simple: connect a database, write queries, build charts and dashboards, share access with your team. That's it.

3. Why Does a BI Tool Even Matter?

Everyone has data. The problem isn't a lack of it - the problem is that only people who can write SQL, or who know the right person to ask, can actually get to it.

Without a BI tool, the typical picture looks like this:

1. An analyst or developer becomes the sole gateway to data;
2. A manager articulates a question in plain language, the developer translates it into SQL;
3. The result gets exported to Excel and sent by email;
4. A week later the data is stale and the whole cycle repeats.

BI tool breaks this cycle. Data becomes accessible, up to date, and understandable - without a middleman, without repetition.

4. Why Superset and Not Tableau or Power BI?

The honest answer: because it's free.

1. Superset - free, ✅ Self-hosted, ✅ Open-source, ✅ SQL editor, Data sources - 40+, Learning curve - Medium;
2. Tableau - paid (~$70/month per user), ❌ Self-hosted, ❌ Open-source, SQL editor - Limited, Data sources - 80+, Learning curve - Low;
3. Power BI - paid (~$10–20/month), ❌ Self-hosted, ❌ Open-source, SQL editor - Limited, Data sources - 100+, Learning curve - Low;

Tableau and Power BI are considerably more powerful in enterprise scenarios. But if you have a team of 25 people, Tableau will cost you at least $20,000 a year - just in licensing fees.

5. What It Looked Like in Practice

Before:

Our project had a PostgreSQL database with an orders table. Managers wanted to see metrics by week, by region, by product. Every single request came through me.

5-10 Slack messages a day. I kept losing focus; they kept waiting for answers. Nobody was happy.

After:

I got Superset up and running in one evening, connected the database, and built a few basic dashboards that I refined over time. Then I gave the team access.

The next day, a manager messaged me: "Wow, I just looked at the customer data myself. This is great!"

Since then, the number of data requests has dropped dramatically. The team checks the dashboards on their own. The data is always current. I write code.

6. How to Get It Running in 20 Minutes

Step 1 - docker-compose.yml

Create a docker-compose.yml file:

Step 2 - Initialization:

Step 3 - Connect PostgreSQL:

Go to `http://localhost:8088` → Settings → Database Connections → + Database

Example of a connection string:

7. Your First Query in SQL Lab

SQL Lab is Superset's built-in SQL editor. Open it, select your database, and write a query:

Results appear immediately in a table. One click turns it into a chart. Another click adds it to a dashboard.

8. Customization

Superset supports custom CSS for dashboards, so you can adapt the look to your company's brand guidelines directly in the interface.

Edit Dashboard → CSS:

9. Roles and Permissions

Apache Superset has a robust built-in role and access management system, which I found quite flexible: managers get read-only access, analysts can edit, developers get full control.

Example of creating a custom role via CLI:

10. Custom Charts

If the built-in visualization types aren't enough, you can always add your own via Apache ECharts or write a plugin in React:

11. An Honest Look at the Downsides

It wouldn't be fair to leave this out:

1. Deployment is complex for non-technical users - without Docker and a basic understanding of DevOps, it won't be easy;
2. Documentation has gaps, especially for advanced use cases;
3. Less corporate support - there's no dedicated support tier like Tableau or Power BI offer;
4. Custom plugins require React knowledge in most cases.

That said, for a development team running PostgreSQL and looking to save time and energy, the trade-off is clearly worth it.

12. Conclusion

I didn't become a business analyst. I simply gave my team the tools they needed to stop depending on me.

Superset isn't magic. It's the right tool in the right place.

One Docker container, a few hours of setup, and data is no longer the exclusive domain of people who can write SQL.

Any developer can become the chef of their data kitchen - you just need the right tools!