3rd way to setup Metabase in production

Easy Way to Install Metabase in Production

Well actually, the Metabase Team already wrote “How to run Metabase in production” where they covered the first 2 official ways.

However, there is a 3rd way and that’s what I’m about to share here.

It’s also self-hosting Metabase but on steroids as it solves lots of pain points in maintaining an open source project.

Based from my experience as a Data Engineer, it’s never fun to upgrade an open-source project.

Instead of me working on my core responsibility to discover insights for the business, I ended up working mostly on fixing the open-source tool.

Enough of the chit-chat! Let’s dive in!

3 steps to setup an automated self-hosted Metabase

  1. Choose your cloud service provider. This is where Plural in step 2 will be deployed. Currently these are the supported providers: AWS, GCP & Azure. Follow this guide to make sure it’s set up correctly.
  2. Install Plural.sh. Plural is an open-source management platform. Once installed, it automatically handles the deployment, maintenance & scaling of the open-source apps installed on this platform which in our case it’s Metabase. Please follow this quick start guide to get it done correctly.
  3. Install Metabase via Plural.sh. Once you have Plural setup, it’s now easy to install open-source apps that are available in their marketplace. Just search for Metabase and install it from there. The installation of support open-source applications in Plural can also be found in the quick start guide.

That’s pretty much it! So easy right?

Benefits of this approach

  • Cost savings. Since this is self-hosted, it means that you can save cost but at the same time enjoying the full benefits of Metabase Cloud Starter. See pricing.
  • Cost savings. In this approach, I bet the Data Engineer can do the setup so no need to hire an FTE which is another way to save cost.
  • Get business insights fast. Since we installed Metabase in Plural, Plural will automatically handle the maintenance upgrades and auto-scaling which means the Data Engineer can focus on discovering and delivering the business insights instead of doing the maintenance.

If you don’t have a Data Engineer but you badly need business insights for your business, please feel free to email me at joey@taleno.digital and you’ll get more cost savings. 🫰