What-If Analysis Skill
Scenario Comparison: Analyze projected metric values under different business assumptions. Model alternative MAQL calculations for measures and compare results across scenarios side by side.
Experimental Feature
This is an experimental feature that is still under active development. Its behavior may change in future releases, or the feature may be removed.
How It Works
The What-If Analysis skill compares projected metric values under different business assumptions by creating alternative MAQL calculations. When activated, the assistant:
- Identifies the base metric and the variable to change
- Creates scenario metrics with different MAQL expressions (e.g., Revenue × 0.9, Revenue × 0.8)
- Uses the Visualization skill to compare scenarios side by side
- With data sharing enabled, interprets results and provides recommendations
- Shows comparisons in a single visualization for easy analysis
The skill helps you understand the impact of different business decisions by modeling alternative scenarios and comparing their projected outcomes. Scenarios must be expressible as metric formulas using arithmetic operations.
Examples
Comparing discount scenarios:
User: "What if we discount by 10% vs 20%?"
Assistant: [Creates scenario metrics: Revenue × 0.9 and Revenue × 0.8, compares in chart, explains differences]Modeling growth assumptions:
User: "Compare revenue if customer growth is 10% vs 15%"
Assistant: [Creates scenario metrics with different growth rates, visualizes comparison]Analyzing cost changes:
User: "What happens to profit if we reduce costs by 20%?"
Assistant: [Creates scenario with adjusted cost metric, compares profit outcomes]Other use cases:
- “Compare revenue if we increase prices by 5% or 10%”
- “What if customer growth is 10% vs 15%?”
- “Show me revenue under different cost scenarios”
- “Compare sales if we add a new region”
Limitations
- Requires the Visualization skill as a dependency
- Scenarios must be expressible as metric formulas (arithmetic operations)
- Base metric must exist or be created first
- Results interpretation requires data sharing to be enabled (when disabled, provides visualization only)
- Complex scenarios may require multiple metric dependencies
- Users need permission to create metrics for scenario metrics