You can create separate workspaces for staging and production, enabling safe iteration on new features and pipeline testing without risking production stability.

​
Why use Staging and Production?

Using separate workspaces for staging and production is a critical best practice for managing reliable systems. This separation allows you to isolate development and testing environments to prevent unintended disruptions to live systems.

​
Staging Workspaces

  • Contain staging data and connection settings
  • Serve as testing environments to validate changes, perform integration testing, and simulate production behavior
  • Allow teams to iterate on new features or pipeline configurations without affecting live systems
Use cases:
  • Testing a new pipeline or query configuration with mock or limited data before deploying it to production
  • Debugging and resolving issues that could arise in production without impacting live users

​
Production Workspaces

  • Use live data and connection settings
  • Serve data to real users, powering your production systems
  • Provide a stable and high-performance environment for your pipelines
Use cases:
  • Delivering production-ready API endpoints to customers or applications
  • Running queries on the full production dataset
  • Monitoring pipeline performance and ensuring real-time data accuracy for end-users

​
Example Workflow

  1. Create Workspaces:
    • Set up two workspaces through app.us.airfold.co
      • One for staging, another for production
    • For example, we may set up staging and production workspaces for sales_calls as follows:

data format options

  1. Develop in Staging:
    • Create a new pipeline in the staging workspace and test it with a subset of data
  2. Validate Results:
    • Ensure that the pipeline produces the expected results
    • Test integration with external systems or mock API calls using staging credentials
  3. Promote to Production:
    • Once validated, deploy the same pipeline to the production workspace using live data and credentials
  4. Monitor Production:
    • Continuously monitor the production workspace for pipeline performance, data accuracy, and API reliability
You may also separate staging and production through branching, as detailed in Version Control.