agentleFS
nisarthnisarth/claude-skills-store/skills/finance/variance-analysis/SKILL.md

variance-analysis

Skill1 starsChanged 7 months agoMITView on GitHub

Decompose financial variances into drivers with narrative explanations and waterfall analysis. Use when analyzing budget vs. actual, period-over-period changes, revenue or expense variances, or preparing variance commentary for leadership. Trigger with "variance analysis", "budget vs actual", "explain the variance", "P&L drivers", or "waterfall analysis".

7 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

8 rules. 0 are common, 8 are distinctive, and 0 are very specific to this project.

  • 0

    Common

    Most of the 5 edited versions we read keep these.

  • 8

    Distinctive

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Create a waterfall (bridge) analysis from base to actual:

    line 74
  2. Distinctive:

    Start with the base amount.

    line 76
  3. Distinctive:

    Add each driver as a step (positive or negative).

    line 77
  4. Distinctive:

    Verify the drivers sum to the total variance.

    line 79
  5. Distinctive:

    Always verify that decomposed drivers sum back to the total variance.

    line 150
  6. Distinctive:

    Use consistent sign convention throughout (favorable = positive, unfavorable = negative, or vice versa).

    line 153
  7. Distinctive:

    When multiple drivers contribute to one line item variance, rank them by magnitude.

    line 155
  8. Distinctive:

    Always include an outlook statement: "Is this variance expected to continue?"

    line 156

“Common” is measured: kept by at least 6 in 10 of the 5 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Variance Analysis

Decompose financial variances into component drivers, quantify each driver, produce narrative explanations suitable for management, and format as waterfall analysis.

Workflow

Step 1: Define the Comparison

Establish the analysis framework:

  • Base period: Budget, prior month, prior year, or forecast.
  • Actual period: The period being analyzed.
  • Scope: Full P&L, specific line items, a department, or a cost center.
  • Materiality threshold: Dollar and percentage thresholds for flagging (e.g., >$25K and >5%).
  • Sign convention: Favorable (F) increases income; Unfavorable (U) decreases income.

Step 2: Calculate Variances

For each line item in scope:

  • Dollar variance: Actual minus base (or base minus actual for expenses where under-spend is favorable).
  • Percentage variance: Dollar variance divided by base amount.
  • Favorable/Unfavorable flag: Apply sign convention.
  • Materiality flag: Mark items exceeding either threshold.

Step 3: Decompose into Drivers

Break material variances into their root causes:

Revenue Variances:

  • Volume: Change in units sold multiplied by base price.
  • Price/Mix: Change in average selling price multiplied by actual volume.
  • New/Lost: Revenue from new customers vs. churned customers.
  • Timing: Revenue pulled forward or pushed out vs. plan.

Expense Variances:

  • Headcount: Change in FTE count multiplied by average cost per head.
  • Rate: Change in cost per unit (salaries, vendor rates) multiplied by actual volume.
  • Volume/Usage: Change in consumption (cloud spend, travel, materials).
  • One-time: Non-recurring items (settlements, write-offs, project costs).
  • Timing: Spend that shifted between periods.

Gross Margin Variances:

  • COGS rate change: Impact of unit cost changes on margin.
  • Revenue mix shift: Impact of product mix on blended margin.
  • Volume leverage: Fixed cost absorption at different volume levels.

Step 4: Build the Waterfall

Create a waterfall (bridge) analysis from base to actual:

  1. Start with the base amount.
  2. Add each driver as a step (positive or negative).
  3. Arrive at the actual amount.
  4. Verify the drivers sum to the total variance.

Step 5: Draft Narrative Commentary

For each material variance, write a management-ready narrative:

  • State the variance in dollars and percentage.
  • Name the driver(s) with quantification.
  • Explain the underlying business reason.
  • Note whether the variance is expected to continue (run-rate) or is one-time.
  • Recommend action if the variance is unfavorable and controllable.

Output Format

## Variance Analysis: [Scope] | [Actual Period] vs. [Base]

### Summary
| Metric             | Base       | Actual     | $ Variance | % Variance | F/U |
|--------------------|-----------|------------|------------|------------|-----|
| Net Revenue        | $10,000K  | $10,800K   | $800K      | 8.0%       | F   |
| Gross Profit       | $6,000K   | $6,200K    | $200K      | 3.3%       | F   |
| Operating Expenses | ($4,500K) | ($4,900K)  | ($400K)    | 8.9%       | U   |
| Operating Income   | $1,500K   | $1,300K    | ($200K)    | (13.3%)    | U   |

### Waterfall: Operating Income
Base (Budget):          $1,500K
  + Revenue volume:     +$500K
  + Revenue price/mix:  +$300K
  - COGS rate increase: -$100K
  - S&M headcount:      -$250K
  - R&D cloud spend:    -$100K
  - G&A one-time legal: -$50K
  - Other:              -$100K (net)
= Actual:               $1,300K

### Material Variance Detail

#### 1. Revenue: +$800K favorable (8.0%)
**Drivers**:
- Volume (+$500K): 12 new enterprise contracts closed vs. 8 planned...
- Price/Mix (+$300K): Higher proportion of premium tier subscriptions...
**Outlook**: Volume trend expected to moderate in Q2; price/mix sustainable.

#### 2. Operating Expenses: ($400K) unfavorable (8.9%)
**Drivers**:
- S&M headcount (-$250K): 3 AEs hired ahead of plan in January...
- R&D cloud infrastructure (-$100K): Migration to new platform...
- G&A legal fees (-$50K): One-time patent filing costs...
**Outlook**: Headcount is run-rate; cloud and legal are one-time.

### Action Items
| # | Item                            | Owner      | Due Date   |
|---|---------------------------------|------------|------------|
| 1 | Review cloud spend trajectory   | VP Eng     | 2026-04-15 |
| 2 | Update S&M headcount forecast   | VP Sales   | 2026-04-10 |

Examples

User: "Explain why we missed operating income by $200K vs. budget" Action: Pull actual vs. budget by line item, identify material variances, decompose into drivers, build the waterfall, and draft narratives.

User: "Revenue bridge from Q4 to Q1" Action: Decompose the quarter-over-quarter revenue change into volume, price/mix, new/lost customer, and timing drivers.

User: "Prepare variance commentary for the board deck" Action: Summarize the top 5 variances with concise driver explanations and forward-looking outlook.

Best Practices

  • Always verify that decomposed drivers sum back to the total variance. If they do not, include a "rounding/other" line.
  • Lead with the story, not the numbers. Management wants to know why, not just what.
  • Distinguish between run-rate and one-time variances; this changes the forecast implications.
  • Use consistent sign convention throughout (favorable = positive, unfavorable = negative, or vice versa).
  • For board-level commentary, keep each variance explanation to 2-3 sentences maximum.
  • When multiple drivers contribute to one line item variance, rank them by magnitude.
  • Always include an outlook statement: "Is this variance expected to continue?"

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.