Building a semantic layer with dbt + Metriql and synchronizing metrics to Tableau & Looker.

[emresemercioglu]

We'll build a semantic layer(aka metrics layer) using dbt + Metriql and synchronize metrics to Tableau and Looker and write a blog to showcase how powerful it's to use the same metrics in multiple BI tools at once. It can be considered as a follow-up post on this piece

Steps
1- Fork https://github.com/metriql/metriql-public-demo and copy the models & seeds from dbt sample project: https://github.com/dbt-labs/jaffle_shop
2- Load this data into a data warehouse using your preferred ETL tool.
3- Define metrics relevant to the dataset using Metriql (you should make use of Metriql's relations
4- Use Metriql's integrations to sync metrics to Tableau and Looker and start running ad-hoc analysis.
5- Prepare dashboards on both Tableau and Looker (you'll link them in your blog later on)
6- Write a blog post/tutorial including screenshots and a short video to explain how you built this semantic layer and compare it to other semantic layer solutions such as LookML.

Tooling

• You can use Github's VSCode (example: https://github.dev/metriql/metriql-public-demo) and mention it as an alternative to Looker IDE.
• If you fork the public demo, it comes with Github Actions that run dbt as part of the continuous integration process. (https://github.com/metriql/metriql-public-demo/actions) You can mention Github Actions as an alternative to Looker's internal scheduler.
• We use CI and CD to deploy the data models to Heroku automatically when the data models change. (Bonus: testing data models as part of the CI process. Talk to @buremba about it before publishing the post)

[mirorac]

Is this still under consideration? I'm interested in learning how to utilize the Tableau API to synchronize metrics that are maintained within a dbt project, whether with or without metriql.