Software measurement is a challenging as well as an essential element of a wholesome and highly efficient software engineering culture. ‘Software metrics’ refers to the collective term underlying the wide range of activities associated with measurement in software engineering. There are numerous taxonomies to organize software metrics (based on entities, type of measure etc), but all of them fundamentally fit into three broad domains, namely resource, process and product metrics. The number of metrics in each of these categories is ever increasing with the result that practitioners are overwhelmed with data as they get quantitative inputs on a variety of attributes. This poses a real problem for decision-makers and analysts as many times it is unclear which attributes should be managed so as to augment the overarching goal of productivity or quality improvement. This paper aspires to address the research issues with respect to ably managing the subject of software metrics in a structured and coordinated manner. The present work deals with the concept of software measurement and metrics and argues that a great degree of simplification is required as, otherwise, the cornucopia of metrics would defeat the very purpose of the measurement approach, which is to quantify the key characteristics of interest in order to extract meaningful inferences about them and their relationships with one another. A cause-analytic model is also proposed, which if validated empirically would help to throw light on certain key metrics that could form the foundation for the further realization of the measurement and metrics approach. To encapsulate, the proposed research adopts a holistic approach in that it encompasses all categories of metrics in its analysis and it takes a macro-perspective in the sense that the sole purpose is to better organize the body of research with respect to software metrics.