"There are no classes in life for beginners: right away you are always asked to deal with what is most difficult."- Rainer Maria Rilke

Test Management

Software testing is the process used to help identify the correctness, completeness and quality of the computer software developed to help support your key business processes. With that in mind I believe that it's important for any organisation regardless of scale to ensure that the testing process is adequately managed. Although adherence to industry recognised "best practice" is always advisable, so too is a degree of pragmatism as a real life project rarely follows the ideal world scenarios on which the best practice approach is usually based. More often than not it's an adaptation of the best practice methods that produce acceptable results for real clients, with real timescales and real customers on the receiving end.

To support this I believe that the testing process should be managed to ensure that it is relevant, effective, adaptable and that it follows the development methodology adopted by your organisation whether that is the V-model, a RAD methodology, or another approach (perhaps developed in-house) that is more appropriate to your organisations style of working, constraints and core values.

Test execution can be a manual process. It can be an automated process. Both of these aspects have their place and by careful management it's possible to find a balance that's right for your circumstances.

Effective test management starts with the acceptance that although it's possible to plan the start date for testing, it isn't quite so straightforward to accurately plan the end date of testing. I accepted this fact of test management life a long time ago. A natural consequence of this acceptance is the realisation that the end date for the testing phase is only derived through a risk assessment and by managing the entire test process based on that risk assessment - testing can then be prioritised to balance the perceived risk and decisions can be made based on expected impact to your business. Minimal surprises. No shocks.

It's a fact that testing is expensive. It's a more disturbing fact that not testing can prove to be considerably more expensive for your business.

Treating the testing process as an engineering discipline like any other in the development lifecycle and ensuring effective, pragmatic management from the beginning can help ensure that your organisation targets the risks and controls its development budget by spending where it's likely to make a difference - in the test lifecycle.