Concrete Strategies

In practice, testing strategies are built up using a mixture of the various approaches described above. For example, you can combine risk-based testing (an analytical approach) with exploratory testing (a reactive strategy element). The following variants are often used in real-world situations:

• Cost-based testing
  The testing techniques you select should be driven mainly by the need to optimize the relationship between costs/time required and the number and complexity of the required test cases. Simple, broad-based tests are preferable to costly, in-depth tests, which should be avoided.
• Risk-based testing
  This approach analyzes data relating to project and product risks, and focuses testing on the highest-risk areas. Risk is the parameter that influences strategic decision the most.
• Model-based approach
  This approach uses abstract models of the test object to derive test cases, define exit criteria, and to measure test coverage. An example of this approach is state transition testing in which a state machine serves as the model. Other examples of models that can be used to draft a testing strategy are static models of fault distribution within the test object, failure rates in use (reliability model), or the frequency of use cases (mission or usage profiles).
• Methodical approach
  This kind of strategy is based on the systematic use of predefined sets of tests or test conditions. For example, a set of common or probable defects, a list of important quality characteristics, or company-wide “look and feel” standards for mobile apps and websites.
• A reuse-based approach bases testing on existing tests and test environments from previous projects. The aim is to set up the testing process quickly and pragmatically.
• A checklist-based approach utilizes lists of defects from previous test cycles¹⁰, lists of potential failures¹¹ and risks, prioritized quality characteristics, and other informal base materials.
• A process- or standards-compliant approach utilize guidelines or recommendations¹² that can be combined in testing “recipes”. Sources can be generally accepted industry-wide or company-internal standards, or other legal stipulations and regulations.
• The directed (or consultative) approach leverages expertise and the “gut instincts” of testing experts, whose attitude to the technologies in use and/or the business domain influences and controls the choice of testing strategy. This kind of testing strategy is primarily determined by the advice, guidance, and instructions from stakeholders, experts and technologists external to the test team or even to the company itself.
• Regression-averse approach
  This kind of testing strategy is motivated by the desire to preserve product performance, and is based strongly on the reuse of existing test cases and test data, extensive test automation, and the repetition of these tests using regression test suites.