Automated Testing as a Prerequisite for Continuous Testing and Layered Test Strategies

In continuous testing practice, automated testing is indispensable because it enables the verification of user stories or application requirements and serves as a quality gate for the delivery pipeline, making it a necessary condition for continuous testing.

Automation skills have become a standard requirement for test engineers; most job postings now include automation expectations, indicating that automation has become a core competency. By using tools, frameworks, and code, automation replaces repetitive manual work, improves efficiency, and reduces the workload of test engineers.

Layered automated testing builds test suites at different abstraction levels to maximize test value. By addressing gaps between layers and focusing on testability and quality risk, teams can prioritize testing efforts, enhance controllability, observability, simplicity, stability, and information richness of the system.

Quality risk refers to the likelihood and impact of software problems. The article references Heinrich's 300∶29∶1 safety law, illustrating that many minor issues precede serious incidents, and stresses the importance of early detection and mitigation of technical risks through layered testing.

The classic testing pyramid model allocates the most effort to unit tests, followed by API tests, with the least effort on UI tests. Over time, investment has shifted toward more API testing and less UI testing, forming a “rugby‑shaped” model where unit testing receives relatively less focus.

Emerging intelligent testing technologies aim to equalize investment across unit, API, and UI automation, with platforms and algorithms generating test scripts, data, execution, and analysis. Although this vision represents the future direction of automated testing, current maturity levels still lag behind the ideal model.

Figure 2‑1: Heinrich's Law

Figure 2‑2: Evolution of Layered Testing Models