Monday, July 22, 2013

Automated metrics from repositories for better project insight

Metrics are important to understand project health.  Every project manager has to present them for  review meetings. Traditionally spreadsheet software was used extensively for  metrics preparation. Project Manager has to do a lot of work to contact team members for status information and then update the  spreadsheets. The metrics are subjective and are delayed.  Most of these spreadsheets also will not be ideal for knowledge management in organization, as these are not archived and available for future reference.This situation has improved with  increased use of   non spreadsheet  tools in the recent times.

Usually tools used for requirements, coding, code review, testing  and change  management and project management,  have  built-in reports. These are useful for understanding the status and issues in a project in a timely manner. These may require customization to conform to the specifics of process followed by the organization.  If the tools are not integrated,  consolidation of metrics requires additional effort and possibly use of spreadsheets again with their inherent advantages and disadvantages. Software analytics is a good alternative when the tools are not integrated.

Open source communities are leveraging tools like MetricsGrimoire  (See sample figure above from Dashboard of Wikimedia Development Community) to mine repositories associated with  configuration management, bug tracking and mailing list for projects and extract useful metrics and present them visually for analysis and action. These metrics are useful  to identify  top contributors, aging of artifacts, velocity of development as well as provide insight into issues.

What is the level of  non spreadsheet tools usage in your organization? How much time is spent in generating the metrics? Is there an opportunity to deploy additional tools, deploy software analytics tools  and develop automation scripts to extract metrics? These are all valid questions which can lead to improved productivity and project success.