The TMMi Standard Reference Model is intended to be generic enough to apply to all test life cycle activities and delivery models but specific enough to cover all anticipated activities and deliverables.
However, it is important to stress that the model needs to be interpreted by the assessment team depending on a number of contexts including:
- The size of the organisational unit being assessed – Is it small or very large?
- Is it in a single location or spread throughout multiple sites in multiple countries?
- These will influence the interpretation of the model as they influence the degree of rigour to achieve compliance For example, It is fundamentally important for a large organisation which is spread over different cities or countries to have a defined test policy that is published, mandated, known and accessible to all. By contract, a much smaller organisation which has one, limited IT department may not have a corporately published test policy. However, the team may be working to a set of organisational goals. If these goals were published, known to all and acted upon it would fulfil the needs of the requirement.
- The industry sector of the organisational unit being assessed – is it in a rigorous industry sector (such as safety critical software) or is it in a consumer-led industry sector?
- These determine the level of rigour, specification and control required of the testing process and varies between different sectors.
- Does it use a single development methodology e.g. v-model only, or does it use a combination of methodologies e.g. Agile plus a waterfall integration and UAT.
- The demonstration of the Process Areas will be very different although the principles will be the same. For example, with a v-model life cycle, test plans and test cases are defined in full detail before development commences but in Agile these may be done in parallel with ‘sprints’.
- Does the organisation only test one type of domain – e.g. Banking? Or does it handle many domains: Retail, Insurance, Government etc.?
- The organisation my have slightly (or very) different processes for each domain.
Ultimately, Is the process fit for purpose in the context of the organisational unit being assessed? The test plan template, for example, may differ in size and content depending on the pragmatic needs of the unit. The assessment team need to balance necessary content with what is pragmatically useful to the organisational unit to achieve institutionalisation and robust repeatability. Does it satisfy in the context of the needs of the Organisational unit?