If you want to extend a dimension with new attributes from another data source, which steps would you perform first?

• A. Create a sandbox and edit an existing logical star
• B. Directly modify the production schema
• C. Install a new tool
• D. Rebuild the entire semantic model

Answer: (A)

The idea is to prototype changes in a safe environment before touching production. Creating a sandbox and editing the existing logical star lets you add the new attributes from the other data source, define how they join to the dimension, and validate mappings, data types, and semantics without risking live reports or metadata. This isolated approach keeps the current model stable while you verify that the integration works as intended. Directly modifying the production schema is risky and can disrupt analyses and downstream mappings. Installing a new tool isn’t necessary for this task, and rebuilding the entire semantic model is unnecessary drama when you can validate incremental changes in a sandbox first.