One of my clients asked me to compare metrics so I built this table.
Regards, Capers Jones
Function Points | Lines of Code | Story Points | Use-Case Points | |
---|---|---|---|---|
Points | Code | Points | Points | |
ISO standard? | Yes | No | No | No |
OMG standard? | Yes | No | No | No |
Professional associations? | Yes | No | No | No |
Formal training? | Yes | No | No | No |
Certification exam? | Yes | No | No | No |
Automated counting? | Yes | Yes | No | No |
Required by governments? | Yes | No | No | No |
Good for productivity? | Yes | No | Yes | No |
Good for quality? | Yes | No | No | No |
Good for estimates? | Yes | No | Yes | No |
Published conversion rules? | Yes | No | No | No |
Accepted by benchmark groups? | Yes | Yes | No | No |
Used for IT projects? | Yes | Yes | Yes | Yes |
Used for web projects? | Yes | Yes | Yes | Yes |
Used for cloud projects? | Yes | Yes | Yes | No |
Used for embedded projects? | Yes | Yes | No | No |
Used for systems software? | Yes | Yes | No | No |
Used for telecom software? | Yes | Yes | No | No |
Used for defense software? | Yes | Yes | No | No |
Productivity Measures | ||||
Activity-based costs? | Yes | No | No | No |
Requirements productivity? | Yes | No | No | No |
Design productivity? | Yes | No | No | No |
Coding productivity? | Yes | Yes | No | No |
Testing productivity? | Yes | Yes | No | No |
Quality assurance productivity? | Yes | No | No | No |
Technical writer productivity? | Yes | No | No | No |
Project management productivity? | Yes | No | No | No |
Net productivity of projects | Yes | Yes | Yes | Yes |
Quality Measures | ||||
Requirements defects? | Yes | No | No | No |
Architecture defects? | Yes | No | No | No |
Design defects? | Yes | No | No | No |
Document defects? | Yes | No | No | No |
Coding defects? | Yes | Yes | No | No |
Bad fix defects? | Yes | Yes | No | No |
Net quality of projects? | Yes | Yes | Yes | Yes |
Annual rate of change in usage? | 20% | -15% | 15% | -5% |
U.S. usage - companies | 2500 | 2000 | 2500 | 500 |
U.S. usage - current projects | 30000 | 20000 | 15000 | 1000 |
U.S. usage - benchmarks | 90000 | 10000 | 5000 | 1500 |
Global usage - companies | 4500 | 3700 | 4000 | 400 |
Global usage - current projects | 65000 | 30000 | 15000 | 600 |
Global usage - benchmarks | 150000 | 15000 | 5000 | 1500 |
TOTAL USAGE - COMPANIES | 7000 | 5700 | 6500 | 900 |
TOTAL USAGE - PROJECTS | 95000 | 50000 | 30000 | 1600 |
TOTAL USAGE - BENCHMARKS | 240000 | 25000 | 10000 | 3000 |
Major benefits | Accuracy | Easy to use | Agile | IBM origins |
Economics | metric | |||
Quality | ||||
Productivity | ||||
IBM origins | ||||
Benchmarks | ||||
Major problems | Slow | Inaccurate | Inaccurate | Inaccurate |
Expensive | Varies by > 500% | Varies by > 400% | Varies by > 200% | |
Penalizes high-level languages | Only works with User Stories | Only works with Use Cases | ||
Government mandates contracts? | Brazil Italy Japan Malaysia South Korea | none | none | none |
Major benchmark groups | CAST COSMIC DCG FiSMA Galorath Gartner IBM IFPUG ISBSG JFPUG Level 4 Malaysian Test Metricas Metri Group Namcook Nesma QPMG QSM Reifer SPR Total Metrics | US Air Force Price Systems Reifer | Construx | IBM (internal) |
About the author:
Capers Jones is CTO of Namcook Analytics, a company that builds advanced risk, quality, and cost estimation tools. This blogpost was originally posted on the Namcook Analytics blog.