This chapter is about the decisions that shape how a team tests inside an iteration and after it: which test types belong where, when end-to-end testing earns its cost, how formal testing and holistic testing trade off against each other, and which regression test approach fits a given delivery context. Its keywords are examinable in their own right: bug bash, feature toggle, hardening iteration, contract testing, user journey.
The trap here is the level of abstraction. Questions ask about the test approach of one project or one iteration, not about an organisation-wide test strategy, so an option phrased as company policy is usually the distractor. The second trap is treating end-to-end testing as a virtue in itself: the syllabus admits it only through risk, feedback time and the maturity of the continuous integration and continuous delivery (CI/CD) pipeline, and it warns that needing a hardening iteration may itself signal quality neglected during ordinary iterations.
Learn the benefits and drawbacks of formal and holistic testing as complementary, never as winner and loser, and learn each regression approach by the mechanism it uses rather than by its label.
Test approach for a project versus test strategy of an organisation: only the first is asked about here.
Holistic testing means systems thinking, multiple perspectives and First Time Right. It is not a synonym for informal or ad hoc testing, and it complements formal testing instead of replacing it.
End-to-end testing is justified by risk, feedback time and pipeline maturity, not by the wish to cover the whole system.
Contract testing, including consumer-driven contract testing, is the answer for microservice integration; end-to-end testing is the expensive fallback.
Feature toggles, canary releases and observability signal the DevOps-oriented regression approach; recognising which of the five approaches a scenario describes is the real task.
A hardening iteration is both a sensible place for end-to-end testing and a warning sign about earlier iterations: both statements are true at once.
A bug bash belongs to the collaborative regression approach; it is teamwide regression, not a substitute for exploratory testing.
What is tested during an iteration versus after it is recorded in the Definition of Done and the test approach, not left to individual judgement.
This chapter is covered most heavily (5 questions) in CTAL-AT — Mock Exam 1 (Fintech, Delivery and Booking Platforms), CTAL-AT — Mock Exam 2 (Telemedicine, Online Learning and Logistics Scenarios), CTAL-AT — Mock Exam 3 (Legacy Cores, Distributed Teams and Multi-Channel Services)