How To Calculate Function Points With Examples In Software Engineering

Calculate software size using the Function Point Analysis (FPA) method with this interactive tool

Comprehensive Guide to Function Point Analysis (FPA) in Software Engineering

Function Point Analysis (FPA) is a structured technique for measuring the functional size of software applications. Developed by Allan Albrecht at IBM in 1979, FPA has become an ISO standard (ISO/IEC 20926:2009) and is widely used for project estimation, benchmarking, and productivity analysis in the software industry.

Why Use Function Points?

• Technology Independent: Measures functionality from the user’s perspective, not technical implementation
• Early Estimation: Can be applied during requirements phase before design or coding begins
• Benchmarking: Enables comparison of productivity across projects and organizations
• Contract Management: Used in fixed-price contracts to define scope objectively
• Quality Metrics: Helps calculate defect density (defects per function point)

The Function Point Calculation Process

FPA follows a systematic 5-step process:

1. Determine the Type of Count:
   • Development Project: Count all functions delivered to the user
   • Enhancement Project: Count only added, changed, or deleted functions
   • Application: Count all functions in the existing application
2. Identify the Counting Scope and Application Boundary:

   The boundary separates the application being measured from external applications and users. Functions crossing this boundary are counted.

3. Count the Data and Transactional Functions:

   Identify and classify five function types based on their complexity (low, average, high):

   Function Type	Description	Complexity Weights
   External Input (EI)	User inputs that update internal logical files	3, 4, 6
   External Output (EO)	Outputs to users containing derived data	4, 5, 7
   External Inquiry (EQ)	Input-output combinations that don’t update ILFs	3, 4, 6
   Internal Logical File (ILF)	Logical groups of user-identifiable data maintained within the boundary	7, 10, 15
   External Interface File (EIF)	Logical groups of user-identifiable data referenced but maintained outside the boundary	5, 7, 10

4. Determine the Value Adjustment Factor (VAF):

   The VAF accounts for 14 general system characteristics (GSCs) that can influence the function point count. Each GSC is rated from 0 (no influence) to 5 (strong influence):

   1. Data communications
   2. Distributed data processing
   3. Performance
   4. Heavily used configuration
   5. Transaction rate
   6. Online data entry
   7. End-user efficiency
   8. Online update
   9. Complex processing
   10. Reusability
   11. Installation ease
   12. Operational ease
   13. Multiple sites
   14. Facilitate change

   The VAF is calculated as: VAF = 0.65 + (0.01 × ∑GSC)

5. Calculate the Adjusted Function Point Count:

   The final formula combines the Unadjusted Function Points (UFP) with the VAF:

   Adjusted FP = UFP × VAF

Function Point Complexity Assessment

Each function type is classified into one of three complexity levels based on specific criteria:

Function Type	Low Complexity	Average Complexity	High Complexity
External Input (EI)	≤4 FDETs, ≤1 FTR	5-15 FDETs, 2 FTRs	>15 FDETs or >2 FTRs
External Output (EO)	≤5 FDETs, ≤1 FTR	6-19 FDETs, 2-3 FTRs	>19 FDETs or >3 FTRs
External Inquiry (EQ)	≤5 FDETs, ≤1 FTR	6-19 FDETs, 2-3 FTRs	>19 FDETs or >3 FTRs
Internal Logical File (ILF)	≤50 DETs, ≤1 RET	51-100 DETs, 2-5 RETs	>100 DETs or >5 RETs
External Interface File (EIF)	≤50 DETs, ≤1 RET	51-100 DETs, 2-5 RETs	>100 DETs or >5 RETs

FDET = File Data Element Type, FTR = File Type Referenced, DET = Data Element Type, RET = Record Element Type

Practical Example: Calculating Function Points for an Inventory System

Let’s walk through a complete example for an inventory management system:

1. Identify Function Types and Complexity

• External Inputs (EI):
  • Add New Product (Average: 1)
  • Update Product Quantity (Low: 1)
  • Process Order (High: 1)
• External Outputs (EO):
  • Generate Low Stock Report (Average: 1)
  • Order Confirmation (Low: 1)
• External Inquiries (EQ):
  • Product Search (Low: 1)
  • Stock Level Check (Low: 1)
• Internal Logical Files (ILF):
  • Products (Average: 1)
  • Orders (Average: 1)
• External Interface Files (EIF):
  • Supplier Catalog (Low: 1)

2. Calculate Unadjusted Function Points (UFP)

Function Type	Low (×3)	Avg (×4)	High (×6)	Subtotal
External Input (EI)	1 × 3 = 3	1 × 4 = 4	1 × 6 = 6	13
External Output (EO)	1 × 4 = 4	1 × 5 = 5	0	9
External Inquiry (EQ)	2 × 3 = 6	0	0	6
Internal Logical File (ILF)	0	2 × 10 = 20	0	20
External Interface File (EIF)	1 × 5 = 5	0	0	5
Total UFP	53

3. Determine Value Adjustment Factor (VAF)

After evaluating the 14 general system characteristics, we determine the total degree of influence (DI) is 38:

VAF = 0.65 + (0.01 × 38) = 1.03

4. Calculate Adjusted Function Points

Adjusted FP = UFP × VAF = 53 × 1.03 ≈ 54.59

Typically rounded to 55 function points

Function Points vs. Other Estimation Techniques

Metric	Function Points	Lines of Code (LOC)	Story Points	Use Case Points
Technology Dependence	Independent	Highly Dependent	Independent	Semi-Dependent
When Applicable	Early requirements phase	Design/coding complete	Agile planning	Use case modeling
Standardization	ISO/IEC 20926:2009	None (varies by language)	Team-specific	None
Best For	Contract bidding, benchmarking	Code productivity analysis	Agile sprint planning	Use case driven projects
Learning Curve	Moderate	Low	Low	Moderate

Industry Benchmarks and Statistics

Function point analysis provides valuable benchmarks for software development projects:

• Productivity Rates:
  • Average: 8-12 function points per person-month
  • High productivity: 15-20 function points per person-month
  • Outsourced projects: 5-8 function points per person-month
• Defect Rates:
  • Industry average: 0.5-1.5 defects per function point
  • CMM Level 5 organizations: 0.1-0.3 defects per function point
• Project Size Classification:

  Project Size	Function Points	Typical Duration	Team Size
  Small	< 100	1-3 months	1-3
  Medium	100-500	3-12 months	3-10
  Large	500-2,000	1-3 years	10-30
  Very Large	> 2,000	3+ years	30+

Common Challenges and Best Practices

While function point analysis is powerful, organizations often face these challenges:

1. Subjectivity in Counting:

   Different counters may produce varying results for the same application.

   Best Practice: Use certified counters and maintain an organizational counting guide with examples.

2. Boundary Definition:

   Unclear application boundaries lead to inconsistent counts.

   Best Practice: Document boundary decisions and review with stakeholders.

3. Complexity Assessment:

   Determining low/average/high complexity can be subjective.

   Best Practice: Use decision tables and examples from past projects.

4. Maintenance Overhead:

   Keeping function point counts updated as applications evolve.

   Best Practice: Implement automated tools and make FPA part of change management.

5. Tool Selection:

   Choosing between manual counting and automated tools.

   Best Practice: Use tools for consistency but validate with manual reviews.

Automated Function Point Analysis Tools

Several commercial and open-source tools can assist with function point counting:

• CAST Software: castsoftware.com – Automated function point counting from source code
• SCOPE by Capers Jones: Commercial tool with extensive benchmarking data
• FPA Workbench: Open-source tool for manual counting with validation rules
• IFPUG Sizing Suite: Official tool from the International Function Point Users Group
• COCOMO II: Integrates function points with cost estimation models

Academic Research and Standards

Function Point Analysis has been extensively studied in academic research. Key resources include:

• International Function Point Users Group (IFPUG) – The primary standards body for function point analysis
• ISO/IEC 20926:2009 – The international standard for function point counting
• NIST Software Metrics – U.S. government research on software measurement including function points
• USC Center for Systems and Software Engineering – Research on function points and COCOMO integration

Function Points in Agile Development

While function points originated in waterfall environments, they can be effectively used in Agile:

1. Backlog Refinement: Use function points to size epics and features during backlog grooming
2. Velocity Tracking: Track function points completed per sprint for more objective velocity measurement
3. Release Planning: Combine with story points for more accurate release forecasting
4. Benchmarking: Compare Agile team productivity (FP/person-month) against industry benchmarks
5. Contracting: Use in fixed-price Agile contracts to define scope objectively

Research from the Agile Alliance shows that teams using function points alongside story points achieve 15-20% better estimation accuracy for large projects.

Future Trends in Function Point Analysis

The practice of function point analysis continues to evolve:

• Automation: AI-powered tools that can count function points directly from requirements documents
• Snapshot Counting: Techniques for quickly estimating function points from existing systems
• Cloud Applications: Adapted counting rules for serverless and microservice architectures
• Integration with DevOps: Real-time function point tracking integrated with CI/CD pipelines
• Cognitive Function Points: Extensions for measuring AI/ML system functionality

The Object Management Group is currently working on standards for next-generation function point analysis that better address modern software architectures.

Conclusion

Function Point Analysis remains one of the most robust methods for measuring software size and estimating project effort. When applied correctly, it provides:

• Objective sizing independent of technology
• Early estimation capability from requirements
• Consistent benchmarking across projects
• Improved contract management
• Better productivity tracking

While the learning curve exists, the long-term benefits of function point analysis make it a valuable skill for software estimators, project managers, and business analysts. The calculator provided in this guide gives you a practical tool to begin applying FPA to your projects immediately.

For organizations serious about software measurement, investing in formal FPA training and certification through IFPUG can provide significant returns in estimation accuracy and project success rates.