Sensitivity Analysis Process

When to Use

• User needs to identify project risk factors
• User asks to perform sensitivity analysis
• User wants to evaluate different scenarios (optimistic, base, conservative)
• User needs to calculate risk-adjusted ROI
• User asks for risk management recommendations

Sensitivity Analysis Process

Step 1: Identify Key Variables

Variable Categories:

Category	Typical Variables	Description
Cost	Development cost, operation cost, other costs	Variables affecting project investment
Benefit	Direct benefit, indirect benefit, strategic benefit	Variables affecting project returns
Time	Go-live date, project duration	Variables affecting cash flow timing
External	Discount rate, inflation rate	Variables affecting capital cost

Identification Methods:

1. Historical project experience
2. Similar project case studies
3. Expert interviews and brainstorming
4. Stakeholder feedback

Step 2: Determine Variable Change Ranges

Range Setting Principles:

Scenario	Change Range	Basis
Optimistic	-10% ~ +10%	Ideal situation
Base	0%	Most likely situation
Conservative	-40% ~ +30%	Risk situation

Typical Change Ranges:

Variable Type	Pessimistic	Base	Optimistic
Cost	+20%~+30%	100%	-10%
Benefit	-30%~-40%	100%	+10%
Time	Delay 3 months	On time	Advance 1 month
External	Unfavorable	Base	Favorable

Step 3: Single-Factor Sensitivity Analysis

Analysis Method: Change one variable at a time while keeping others constant, observe impact on financial indicators.

Analysis Steps:

1. Select variable to analyze
2. Set variable change range
3. Calculate financial indicators at different change levels
4. Record analysis results

Output Template:

Variable Change	Financial Indicator 1	Financial Indicator 2	...
-30%
-20%
-10%
Base
+10%
+20%
+30%

Step 4: Calculate Sensitivity Coefficient

Formula:

Sensitivity Coefficient = (Financial Indicator Change Rate) / (Variable Change Rate)

Example:
Sensitivity = (NPV Change Rate) / (Variable Change Rate)
            = (New NPV - Original NPV) / Original NPV / Variable Change Rate

Sensitivity Levels:

Sensitivity Coefficient (Absolute)	Level	Color
> 0.8	High	🔴 Red
0.5 ~ 0.8	Medium-High	🟡 Yellow
0.3 ~ 0.5	Medium	🟢 Green
< 0.3	Low	⚪ Gray

Step 5: Sensitivity Ranking

Ranking Method: Sort by absolute value of sensitivity coefficient (descending)

Output Formats:

1. Sensitivity ranking table
2. Tornado diagram
3. Sensitivity matrix

Step 6: Build Analysis Scenarios

Scenario Settings:

Scenario	Probability	Characteristics	Purpose
Optimistic	20%	All favorable factors occur	Upper bound assessment
Base	60%	Most likely normal situation	Primary decision basis
Conservative	20%	Major risk factors occur	Lower bound assessment

Scenario Parameter Settings:

Parameter	Optimistic	Base	Conservative
Cost adjustment	-10%	100%	+20%
Benefit adjustment	+10%	100%	-30%
Time adjustment	Advance 1 month	On time	Delay 2 months

Step 7: Scenario Analysis Calculation

Calculation Content:

1. Cost and benefit under each scenario
2. Financial indicators under each scenario (ROI, NPV, IRR)
3. Expected financial indicators (probability-weighted)

Expected Financial Indicator Calculation:

Expected Value = Σ(Scenario Value × Probability)

Example:
Expected NPV = NPV_optimistic×P_optimistic + NPV_base×P_base + NPV_conservative×P_conservative

Step 8: Risk-Adjusted ROI

Risk Adjustment Methods:

1. Certainty Equivalent Method:

   Risk-Adjusted Benefit = Expected Benefit × Risk Adjustment Factor

2. Risk-Adjusted Discount Rate Method:

   Risk-Adjusted Discount Rate = Base Discount Rate + Risk Premium

3. Probability-Weighted Method:

   Risk-Adjusted ROI = (Risk-Adjusted Benefit - Risk-Adjusted Cost) / Risk-Adjusted Cost

Risk Adjustment Factors:

Risk Level	Risk Adjustment Factor	Applicable Situation
Low risk	0.95	Mature technology, clear requirements
Low-Medium risk	0.90	Standard projects
Medium risk	0.85	General projects
Medium-High risk	0.80	Innovative projects
High risk	0.75	Breakthrough projects

Step 9: Develop Risk Management Recommendations

Measures for High-Sensitivity Variables:

Sensitive Variable	Risk Measure	Responsible Party
Strategic benefit	Establish tracking mechanism, regular assessment	PMO
Indirect benefit	Data quality monitoring, process optimization	Business department
Direct benefit	User training, feature optimization	Product team

General Risk Management Recommendations:

1. Establish risk monitoring mechanism
2. Develop risk response plans
3. Regular risk assessment
4. Dynamic strategy adjustment

Step 10: Compile Analysis Report

Report Contents:

1. Sensitivity variable identification results
2. Single-factor sensitivity analysis results
3. Sensitivity ranking and tornado diagram
4. Scenario analysis results
5. Risk-adjusted ROI
6. Risk management recommendations
7. Conclusions and investment recommendations

Step 11: Review and Approval

Review Checklist:

• [ ] Variable identification complete
• [ ] Change ranges reasonable
• [ ] Calculations correct
• [ ] Sensitivity ranking correct
• [ ] Scenario settings reasonable
• [ ] Probability estimates reasonable
• [ ] Risk adjustment method reasonable
• [ ] Risk management recommendations feasible
• [ ] Conclusions consistent with data

Approval Process: Preparer → Finance Lead Review → Project Sponsor Approval

Key Formulas

Sensitivity Coefficient

Sensitivity Coefficient = (Financial Indicator Change Rate) / (Variable Change Rate)

Where:
- Financial Indicator Change Rate = (New Value - Original Value) / Original Value
- Variable Change Rate = (New Variable Value - Original Variable Value) / Original Variable Value

Expected Financial Indicator

Expected Value = Σ(Scenario Value × Probability)

Example:
Expected NPV = NPV_optimistic×P_optimistic + NPV_base×P_base + NPV_conservative×P_conservative

Risk-Adjusted ROI

Risk-Adjusted ROI = (Risk-Adjusted Benefit - Risk-Adjusted Cost) / Risk-Adjusted Cost × 100%

Where:
- Risk-Adjusted Benefit = Expected Benefit × Risk Adjustment Factor
- Risk-Adjusted Cost = Expected Cost × (1 + Cost Risk Premium)

Output Templates

Sensitivity Analysis Table

Variable	Change Range	NPV Change	IRR Change	Sensitivity Coefficient	Sensitivity Level
Variable A	-30%~+30%
Variable B	-30%~+30%

Scenario Analysis Table

Scenario	Probability	Cost	Benefit	ROI	NPV	IRR
Optimistic	20%
Base	60%
Conservative	20%
Expected	100%

Risk-Adjusted Analysis Table

Risk Level	Risk Adjustment Factor	Risk-Adjusted Benefit	Risk-Adjusted Cost	Risk-Adjusted ROI
Low risk	0.95
Medium risk	0.85
High risk	0.75

Quality Control

Data Quality Requirements

• Variable identification must be comprehensive, covering all key factors
• Change range settings must have basis
• Scenario probability estimates must be reasonable

Calculation Verification Requirements

• Sensitivity coefficient calculations must be verified
• Scenario analysis calculations must be cross-verified
• Risk-adjusted calculations must be reasonableness-checked

Related Documents

• Sensitivity Variable Analysis (SYS-PI-BC-011)
• Scenario Analysis (SYS-PI-BC-012)
• Risk-Adjusted ROI (SYS-PI-BC-013)
• NPV Analysis (SYS-PI-BC-009)
• IRR Analysis (SYS-PI-BC-010)