Data visualization for webapps?

Question

I have a Postgres database that feeds into a webapp. I would like my users to be able to run set inquiries on the database to get real-time reports and export data as .csv files. They would not be able to edit any of the data. The database currently sits behind a node.js server, and is not exposed to the public. I found Google Charts, and it seems like a good solution. Does anyone have any experience with Google Charts, or something else that can create these reports on demand? Our frontend is in React, if that is helpful.

I am not very technical, but am proficient with SQL and have a developer working for my company.

Thank you!

Answer 1

I build for one of our projects various charts with chart.js (https://www.chartjs.org/). It is a good library for cool looking charts!

At some point we changed the charts for internal use to redash (https://redash.io/) because we needed more and more reports and also caching. You can install redash on your own server that has access to the corresponding database. Then you can simply build your SQL queries in the Redash GUI. You can save them in different dashboards. User management also included. I'm still really excited! We also have a Postgres DB.

The self hosted version is free (open source version): https://redash.io/help/open-source/setup/

Answer 2

A lot depends on what you're looking for in a charting tool. Unfortunately, there's a really tricky tension between usability and expressiveness when it comes to visualization toolkits, meaning that if something is easy to set up, you probably won't be able to do much with it. If their standard plots are what you need, Google Charts are a great choice. If you need something different, or need custom interactivity / control over how data is handled, you'll probably need to build something from scratch with d3.js. There are also some things in the middle—maybe Vega lite has the balance that you're looking for?

Lisa Charlotte Rost has written some excellent blog posts over the years, creating the same chart with 27 different tools, and giving some pros and cons about her experience with each—I'd suggest reading her posts to get a sense for what's out there.

Disclaimer: I built two research tools in this space (Hanpuku and Resonant Lab) that ended up discovering more about why this is a hard problem than actually solving it, so if I sound jaded, that's why.