Features and Aspects That Shouldn’t Be Tested

Features and Aspects That Shouldn’t Be Tested

Each test plan contains features and aspects of a software application that should be left untested. The document has a specific section intended to unambiguously pre-determine which things should not be covered with test cases prepared by the testing team. The number of objects declared as “not subject to testing” may include the following:

Functionality which implementation is postponed until subsequent incremental versions of the product are ready.

Computer system configurations that cannot be checked due to a lack of the necessary equipment. In some cases, computer system configurations appear to be so expensive that they cannot be duplicated in test lab environments. If you cannot reproduce the operating environment, but at the same time you can simulate it, you need to report this in the “Approach” section of your test plan. If you cannot run tests in the operating environment, this issue should be documented in the section that discusses “risks.”

Be informed that various technical analyses are commonly performed within a specialized software testing lab that has all the necessary equipment to achieve desired code coverage. By the way, outsourcing your testing to overseas partners you manage to avoid author bias without sacrificing the product’s quality.  

Combinations of settings or hardware configurations that cannot be tested during the time allotted for the software to enter the market. If you decide to narrow the scope of testing for a particular feature or aspect, you need to discuss these restrictions and justify them in this section.

At the final stages of the project, when, as a rule, pressure is exerted on the testing group to make haste to accomplish  the work, it is very useful to have at your disposal a document that states what should be tested and what – should not. In the heat of pre-final fever, people tend to forget which agreements have been reached – it is quite good to rely on this section of the test plan when answering questions about why a particular feature or aspect is not to be covered by tests. Of course, if the problem arises in an area that has not been tested in your laboratory, it must be eliminated and verified after release of the next version of the software product. Since the cost of repairing defects after the release of a software product is very high, it is important to obtain an accurate assessment of the risks as far as possible before deciding not to test a particular feature or configuration.