Are Test Plans Important to a Project's Success?
“I love writing test cases...said no one ever." This statement captures an often unsaid pain point in the testing community. Writing test cases is somehow considered important, but there aren’t ardent lovers of this activity. Doesn’t the concept of a test plan somewhat follow the same dilemma?
It is prudent to understand the difference between a plan and planning. As Wikipedia defines it, planning is the process of thinking about and organizing the activities required to achieve a desired goal. A plan is typically any diagram or list of steps with timing and resources used to achieve an objective. So a plan is a written artifact resulting from the activity of planning. In other words, a plan represents a static situation; planning is dynamic.
Plans can be thought of akin to a child's toy lying idle; planning is more like repairing the toy while using it to keep the child interested. Given the extent to which software changes during the course of its development, does a static test plan—usually written during the early stages of development—serve any meaningful need?
The authors of the book How Google Tests Software note, "As test documentation goes, test plans actually have the briefest actual lifespan of any testing artefact. Test plans are the first testing artefacts created and the first one to die of neglect."
One relevant question here is: Who actually consumes test plans? The test team is of course the consumer of this document, and the usage depends upon the context. Depending upon the organization's hierarchy and bureaucratic needs, people in higher management are usually kept informed of the test plan. In some cases, test plans are listed as official artifacts to be shared with customers and partners as a part of the initial agreement.
Some other questions that could help decide the importance of the test plan include:
- Does a test plan really drive the testing strategy during execution?
- Does the test team see any value in editing the test plan regularly after it is created?
- Do the non-test consumers of the test plan refer back to it after (and if) they have gone through it?
- Does the test plan serve any value after the release is over?
If the answer to these questions is an unconditional no, then a test plan's role in contributing to the success of a project is obviously in doubt. If the answer is even partially affirmative, then it may be worthwhile to consider using the 10 Minute Test Plan to make things leaner.
Test plan documents may not always be as valuable as the continuous act of planning. Performing the testing for a project without a test plan document may not be the norm in most cases, but putting in some conscious thinking upfront to assess the real need of a test plan helps avoid its being treated just like a placebo.
How do you use test plans?