Installs into .claude/skills of the current project.
Are you the author of Variance?
Add the live security badge to your README. It updates with every re-scan.
[](https://www.skillsdirectory.com/skills/bromso-variance)
---
description: Compare actual financial results against budget or forecast
---
# Variance
Perform variance analysis for "$ARGUMENTS". Compare actual results to budget, forecast, or prior period and explain the differences.
## Prerequisites
None — utility skill, run anytime. More effective when financial data is available from Phase 7 (statements) or Phase 4 (trial balance).
## Process
1. **Establish comparison basis:**
- What are we comparing? (actual vs. budget, actual vs. forecast, actual vs. prior period)
- What level of detail? (total entity, department, account, line item)
- What period? (month, quarter, year)
2. **Calculate variances:**
- For each line item, compute:
- **Dollar variance** — actual minus budget/forecast
- **Percentage variance** — dollar variance divided by budget/forecast
- Classify as favorable (F) or unfavorable (U)
- Highlight variances exceeding materiality thresholds
3. **Analyze root causes:**
- For each significant variance, identify the driver:
- Volume variance (more or fewer units)
- Price variance (higher or lower unit cost/price)
- Mix variance (different product/service mix)
- Timing variance (recognized in a different period)
- One-time items (non-recurring events)
- Quantify each driver where possible
4. **Write the artifact** to `.metapowers/accounting/$ARGUMENTS/variance.md` with sections:
- **Variance Summary** — high-level overview of actual vs. budget with key metrics
- **Detailed Variance Table** — line-by-line comparison with dollar and percentage variances
- **Root Cause Analysis** — explanation of significant variances
- **Recommendations** — actions to address unfavorable variances or capitalize on favorable ones
## Output
The variance analysis written to `.metapowers/accounting/$ARGUMENTS/variance.md`. Present a summary highlighting:
- Top favorable and unfavorable variances
- Key root causes identified
- Recommended actions