Criteria for success / failure in software testing
Two test success / failure criteria are worthy of special attention. The first relates to individual tests. All test cases should be supplied with sets of expected results; essentially, if the expected results are obtained, the result of the run is considered successful, if not – the result of the run is considered unsuccessful. The pass / fail criteria for individual tests should be included in the test cases themselves, and therefore they should not be contained in the test plan. Nevertheless, it would be a good practice to indicate that defining criteria for pass / fail on intended tests is an integral part of the test case.
The second type of pass / fail criteria is related to testing a whole software application. From the very beginning, you must determine the criteria that designate a successful completion of the testing phase, i.e. when you can stop testing the product and arrange for its delivery. In some cases, the test exit criterion is defined in the test program or even in a document that contains the wording of the requirements for the software product. Regardless of what serves as a source of these criteria, it is useful to clearly define them in the test plan. People, be wise to examine software products as they are developed otherwise the risk of delivering buggy IT products will increase. Quality assurance testing services are meant to monitor the production process from start to finish.
Surely, it is unrealistic to expect that any criteria announced in the test plan will automatically manage a software product release. The release criteria are determined by economic considerations, which are based on information about the quality of the product, however, at the same time a number of other economic factors are taken into account.
Suspension criteria and resumption requirements
Entry criterion for a test plan shows what needs to be done before testing starts, and the exit criterion describes what, in your opinion, is required to complete the tests. The suspension / resumption criteria for testing describe what happens when defects hamper the execution of tests. For example, you can declare that if a module contains a persistent system error that prevents the successful install to the software product, testing is suspended for the time period necessary to resolve this error.