Variance Analysis

Execution Steps

Step 1: Identify Analysis Scope

Determine the comparison being requested:

Analysis Types:

• Budget vs. Actual: Compare planned budget against actual results
• Month-over-Month (MoM): Compare current month to prior month
• Quarter-over-Quarter (QoQ): Compare current quarter to prior quarter
• Year-over-Year (YoY): Compare to same period last year
• Forecast vs. Actual: Compare forecasted figures to actual results
• Standard vs. Actual: Compare standard costs/rates to actual (for unit economics)

Scope:

• Full P&L variance
• Specific line item (e.g., marketing spend, salary costs)
• Revenue only (by type, by client, by ICP)
• Cost center or department level

If unclear, ask the user for the comparison type, period, and level of detail needed.

Step 2: Gather Comparison Data

Collect both sides of the comparison:

Actuals:

• Search workspace: Use Glob for *actual*, *financial*, *P&L*, *revenue*, *.xlsx
• Use Grep for revenue, expense, and financial data
• Extract actual figures by line item and period

Budget/Forecast/Prior Period:

• Search workspace: Use Glob for *budget*, *forecast*, *plan*, *target*
• Extract comparative figures for the same line items

If no structured data exists, ask the user to provide:

• Actual figures for the analysis period
• Budget, forecast, or prior period figures for comparison
• Any known context for significant changes

Step 3: Calculate Variances

For each line item, compute:

Calculation Rules:

• Revenue variance: Actual > Budget = Favorable
• Expense variance: Actual < Budget = Favorable
• Margin variance: Actual margin > Budget margin = Favorable
• Use absolute and percentage variance — both matter for different reasons

Aggregation:

• Calculate sub-totals for each category (revenue, COGS, operating expenses)
• Calculate overall impact on operating profit and net profit
• Show the cascade: How does each variance flow through to the bottom line?

Step 4: Benchmark Against Unit Economics (Layaa AI Mode)

Compare key metrics against Layaa AI's established benchmarks:

Apply forecast methodology thresholds from domain-references/revenue-ops/forecast-methodology.md:

• Current month forecast accuracy target: within ±5%
• Next month forecast accuracy target: within ±10%
• Variances exceeding these thresholds require root cause analysis

Step 5: Classify Variances

Categorize each variance:

By Favorability:

• Favorable: Improves profitability (higher revenue or lower costs than planned)
• Unfavorable: Reduces profitability (lower revenue or higher costs than planned)

By Controllability:

• Controllable: Within management's ability to influence (staffing decisions, marketing spend, pricing)
• Partially Controllable: Influenced by external factors but manageable (client churn, conversion rates)
• Uncontrollable: External forces (regulatory changes, market conditions, currency)

By Nature:

• Price Variance: Actual price/rate different from budgeted (e.g., average deal size higher/lower)
• Volume Variance: Actual volume different from budgeted (e.g., fewer/more deals closed)
• Mix Variance: Different product/service mix than planned (e.g., more Starter, fewer Growth retainers)
• Efficiency Variance: More/less resources used than planned per unit of output
• Timing Variance: Revenue/expense in different period than planned (not a true variance)

Step 6: Root Cause Analysis for Material Variances

For any variance exceeding 5% (materiality threshold), perform structured root cause analysis:

5-Why Analysis Framework (from Layaa AI forecast methodology):

1. Why did the variance occur? (Surface-level cause)
2. Why did that happen? (Contributing factor)
3. Why did that factor arise? (Process or system issue)
4. Why was that not prevented? (Control gap)
5. Why does that control gap exist? (Root cause)

Example — Revenue Below Budget:

1. Why? Revenue was 15% below budget
2. Why? Fewer deals closed (3 vs. 5 budgeted)
3. Why? Two proposals were delayed — clients pushed decisions to next quarter
4. Why? Proposals did not address client-specific concerns raised in discovery
5. Why? Discovery process did not capture all decision criteria → Root cause: Discovery template needs revision

Document the 5-Why chain for each material variance.

Step 7: Apply 5-Why Analysis Framework (Layaa AI Mode)

For Layaa AI, focus root cause analysis on these common variance drivers:

Revenue Variances:

• Deal size: Are implementation fees averaging 2.5L or deviating?
• Conversion rates: MQL→SQL (25%), SQL→Proposal (60%), Proposal→Won (35%) — which stage is off?
• Deal mix: Ratio of Starter vs. Growth vs. Enterprise retainers
• Timing: Deals slipping across period boundaries
• Client concentration: Over-reliance on one ICP segment

Cost Variances:

• Headcount: Actual vs. planned team size
• Software/tool costs: New tools added or pricing changes
• Marketing spend: Campaign costs vs. budget allocation
• Delivery costs: More effort than estimated on client projects
• Subcontractor/freelancer costs: Outsourcing vs. in-house delivery

Step 8: Generate Recommendations

For each unfavorable variance, provide actionable recommendations:

Revenue Improvement:

• If conversion below benchmark → Specific stage-by-stage interventions
• If deal size below target → Pricing review, upsell strategy, service packaging changes
• If mix unfavorable → ICP targeting adjustments, sales training on higher-tier packages

Cost Reduction:

• If expense growth > revenue growth → Identify discretionary cuts
• If delivery costs above estimate → Scope management, template/process improvements
• If CAC above 15k → Channel optimization, lead quality improvement

Forecast Improvement:

• If forecast accuracy outside threshold → Identify systematic bias (optimism/pessimism)
• If timing variances dominate → Improve deal stage definitions and close date estimation

Step 9: Update Forecast if Systematic Bias Detected

If variance analysis reveals consistent patterns:

• Revenue consistently above/below by >10% → Adjust forecast methodology assumptions
• Costs consistently above by >5% → Revise cost budgets for remaining periods
• Conversion rates consistently different from benchmarks → Update benchmark rates
• Seasonal patterns identified → Incorporate seasonality into future forecasts

Document recommended forecast adjustments with justification.

Output Format

# Variance Analysis — [Period]
**Analysis Type:** [Budget vs. Actual / MoM / QoQ / YoY / Forecast vs. Actual]
**Period:** [analysis period]
**Comparison:** [what is being compared]
**Prepared:** [date]

## Executive Summary
- **Total Revenue Variance:** [amount] ([%]) — [Favorable/Unfavorable]
- **Total Expense Variance:** [amount] ([%]) — [Favorable/Unfavorable]
- **Net Profit Impact:** [amount] ([%]) — [Favorable/Unfavorable]
- **Material Variances:** [count] items exceeding 5% threshold
- **Key Driver:** [one-sentence summary of the most impactful variance]

## Detailed Variance Report

### Revenue Variances
| Line Item | Budget/Prior | Actual | Variance ($) | Variance (%) | F/U | Classification |
|-----------|-------------|--------|-------------|-------------|-----|---------------|
| Implementation Revenue | [amt] | [amt] | [amt] | [%] | [F/U] | [Price/Volume/Mix] |
| Retainer Revenue | [amt] | [amt] | [amt] | [%] | [F/U] | [Price/Volume/Mix] |
| Training Revenue | [amt] | [amt] | [amt] | [%] | [F/U] | [Price/Volume/Mix] |
| **Total Revenue** | **[amt]** | **[amt]** | **[amt]** | **[%]** | **[F/U]** | |

### Expense Variances
| Line Item | Budget/Prior | Actual | Variance ($) | Variance (%) | F/U | Controllability |
|-----------|-------------|--------|-------------|-------------|-----|-----------------|
| [expense] | [amt] | [amt] | [amt] | [%] | [F/U] | [C/PC/UC] |
| **Total Expenses** | **[amt]** | **[amt]** | **[amt]** | **[%]** | **[F/U]** | |

### Margin Analysis
| Metric | Budget/Prior | Actual | Variance (pp) | Status |
|--------|-------------|--------|-------------|--------|
| Gross Margin | [%] | [%] | [+/- pp] | [vs. 60-80% target] |
| Operating Margin | [%] | [%] | [+/- pp] | [status] |
| Net Margin | [%] | [%] | [+/- pp] | [status] |

## Unit Economics Check (Layaa AI)
| Metric | Target | Actual | Variance | Action |
|--------|--------|--------|----------|--------|
| Gross Margin | 60-80% | [%] | [pp] | [action if off-target] |
| CAC | <15k | [amount] | [amount] | [action if off-target] |
| Avg Deal Size | ~2.5L | [amount] | [amount] | [action if off-target] |

## Root Cause Analysis — Material Variances

### Variance 1: [Description] ([amount], [%])
**5-Why Analysis:**
1. Why? [surface cause]
2. Why? [contributing factor]
3. Why? [process issue]
4. Why? [control gap]
5. Why? [root cause]

**Root Cause:** [concise statement]
**Recommendation:** [specific action]

### Variance 2: [Description]
[repeat format]

## Variance Waterfall

Budget Net Profit: [amount]

• Favorable Revenue: +[amount]

• Unfavorable Revenue: -[amount]

• Favorable Expenses: +[amount]

• Unfavorable Expenses: -[amount] = Actual Net Profit: [amount]

## Recommendations
| # | Action | Expected Impact | Priority | Owner |
|---|--------|----------------|----------|-------|
| 1 | [action] | [impact on variance] | [High/Med/Low] | [name] |
| 2 | [action] | [impact] | [priority] | [name] |
| 3 | [action] | [impact] | [priority] | [name] |

## Forecast Adjustments (if applicable)
| Item | Current Forecast | Recommended Adjustment | Justification |
|------|-----------------|----------------------|---------------|
| [item] | [amount] | [new amount] | [why] |

## Assumptions and Notes
- Materiality threshold: 5% of line item
- [other assumptions]

What Makes This Different from Generic Variance Analysis

• Benchmarks against Layaa AI's unit economics targets (gross margin 60-80%, CAC <15k)
• Uses forecast methodology accuracy thresholds (±5% current month, ±10% next month)
• Applies the 5-Why root cause analysis framework from Layaa AI's forecast methodology
• Decomposes revenue variances by Layaa AI's revenue types (implementation, retainer, training)
• Validates conversion rates against Layaa AI's funnel benchmarks (25% / 60% / 35%)
• Classifies deal mix variances by retainer tier (Starter/Growth/Enterprise)
• Recommends forecast adjustments when systematic bias is detected
• Understands pricing engine context for deal size variance analysis