Looker
Looker is a business intelligence platform. It provides a data exploration and dashboarding interface for users, an IDE for data modelers, and rich embedding and API features for developers.
Pair Team has a ➡️ Looker Cloud Instance (not to be confused with Looker Studio).
Users and Use Cases
- Care Team: as a complement to Arc operational user
- Clinical Area: clinical data exploration, Pop Health initiatives executive user
- Leadership: Critical business metrics, KPIs executive user
- Product: Data analysis, data quality, measure project success analytical user
Properly identifying users and their use cases is the key to a successful Looker data model and dashboards creation.
Looker concepts
-
Familiarize yourself with LookML terms and concepts: models, dimensions, measures, explores, dashboards, looks and views.
-
Learn how to create Looks, Dashboards and organize them in folders
- Saving and Editing Looks
- Creating User Defined Dashboards
- adding filters that affect multiple tiles
- difference with LookML Dasboards
- Learn how to share content
Alerts can be useful for Product to get Slack notifications when some events ocurrs e.g. a message sent an hour ago never received a status update from Twilio
Advanced concepts
Looker development
- Understand Development Mode and Production Mode
- Learn LookML
- Familiarize with the IDE
- Check how Looker does Version Control
- See how to create Views from scratch or from the database
Development workflow
- Enable Development Mode
- Pull from production to update your personal branch
- Make the required changes and test the changes
warning
User Defined Dashboards are not version controlled. That means that any change made to them will be immediately visible to all users.
- Push the changes to your personal branch
warning
Make sure you run the content validator before pushing your changes. This wil help you identify the impact of your changes to existing looks and dashboards.
- Deploy the changes to production
Common development workflows
| Scenario | Branch | Comments |
|---|---|---|
| Single developer | personal branch | Commit to personal branch. Other developers have read-only access to review. |
| Multiple developers | dedicated branch | A branch is created for a particular feature. Multiple developers can collaborate. |
Check how to switch between branches.
Useful tools
- SQL Runner: run raw SQL queries against Looker connections
- Content Validator: validate your changes before pushing them to production
Other resources
Pair Team data model (WIP)
-
Data Sources:
- Arc: Application data from the Postgres DB read replica
- Snowflake: Datawarehouse with data analysis results, Elation / Zus replicas
-
Pair Team models (they are being reworked):
- pairteam
- ecm
- ...