Benefits of the COSMIC method
Size is the most significant factor that determines software effort, cost and timelines. Effective size measurement leads to better metrics, estimation and greater certainty and control. You can learn the basics of COSMIC in a day and reap the rewards for the rest of your career!
The COSMIC method is:
- designed according to basic software engineering principles,
- completely ‘open’ with all documentation available for free download,
- applicable to any kind of software in any layer of a software architecture,
- applicable at any level of granularity or decomposition from a whole application down to its smallest components, including user stories,
- independent of technology and development method,
- governed by an international group of software metrics experts,
- is an International Standard (ISO 19761) and conformant to the ISO 14143/1 standard on the principles of functional size measurement,
- can be used as the basis for metrics on resources, quality and performance/benchmarking.
Compared with Story Points (or other effort-based techniques)
COSMIC sizes are:
- consistent across teams and organisations
- objective, suitable for contratural agreements and not game-able
- understandable by users and technicians
- very strongly correlated to the effort needed to design, build, test and maintain software
Evolved, Mature and Stable
The COSMIC method:
- was evolved from the earlier functional size measurement methods
- was designed to overcome the shortcomings and limitations of those earlier techniques
- has an open-ended measurement scale that is conformant with measurement theory (hence all mathematical manipulations of COSMIC size measurements are valid)
- is thus a pure measure of functional size
- is mature, completely stable, due to its basic design principles
- can measure functionality that may arise from requirements first expressed as ‘non-functional’
- does not require knowledge of all requirements for a sprint or iteration, so is highly suited to Agile
Open and Independent
The COSMIC method is supported by:
- comprehensive documentation; the Measurement Manual has been translated into many languages,
- guidelines that describe how to apply the method to specific types of software, e.g. data warehouse or SOA software, or for specific project management approaches, e.g. Agile methods,
- case studies for business application and real-time software,
- tools for collecting and reporting measurement data,
- tools for analysing requirements for automating size measurement,
- comprehensive benchmark data available via isbsg.org. Many measurements of software projects in different domains have shown an excellent correlation of COSMIC-measured sizes with project effort,
- vendor services of commercial partners, including suppliers of training, consultancy, estimating tools, etc,
- certification exams for applying the method in general or in the early stages,
- guidelines for assuring the accuracy and the comparability of measurements,
- documented variants for approximate COSMIC sizing that can be used early in the life of a project when all the details for the requirements have not yet been established, or for quick size measurement (see 1),
- active user groups on Linkedin (‘COSMIC Users Group’) and Twitter (@COSMIC_FSM),
The COSMIC method is being used successfully around the world for project performance measurement, for estimating, project scope control etc.
The COSMIC method is also being extensively studied by the academic research community. Notable amongst these are several approaches to automating COSMIC size measurement from, e.g. requirements held in UML and from executing programs.
The COSMIC Knowledge Base
The Knowledge Base has a large library of research and conference papers
1 Early Software Sizing with COSMIC, Practitioners and Experts Guides
2 Guideline on how to convert ‘First Generation’ Function Point sizes to COSMIC sizes
3 Recent research (see 1) has demonstrated that COSMIC sizes correlate very well with sizes of the same software measured by the MkII FP method. The size scale of the latter method was calibrated by relating sizes to project effort. The good correlation implies that the COSMIC size scale should be suitable for the purposes for which it was designed (as a component of measuring project performance and as input for project effort estimation), even though the size method was not calibrated against effort.
4 A. Dasgupta, C. Gencel, C, Symons, ‘A process to improve the accuracy of MkII FP to COSMIC size conversions: insights into the COSMIC method design assumptions’, IWSM 2015, Krakow, Poland.