Improving Software Testing Efficiency: From Test Strategy to CI/CD

Yang Xiaohui, former chief test expert at Huawei, shares her 18‑year experience in software testing, emphasizing that testing can significantly improve R&D efficiency when guided by well‑defined strategies rather than blind automation.

She notes that many R&D managers prioritize schedule, cost, and quality, and that testing efficiency begins with a solid test strategy based on risk assessment, which reduces unnecessary tests and discovers critical issues early.

Risk assessment evaluates failure probability and impact, assigning scores using historical data, design complexity, change frequency, and business impact, then allocates test effort accordingly, focusing high‑risk items early and using diverse test techniques.

Beyond strategy, requirement‑based testing helps avoid redundant test cases, while automation should serve quality assurance and, when necessary, improve efficiency for regressions caused by new features.

In the second stage, testing supports development efficiency through automation, tooling, Test‑Driven Development (TDD), and fast defect localization. Automated logging and environment instrumentation enable rapid pinpointing of issues without full reproduction.

The third stage targets release efficiency via automation, Continuous Integration (CI), and Continuous Delivery (CD). CI automates build, test, packaging, and deployment, while CD adds automated UI smoke tests, performance checks, multi‑environment deployment, and rollback mechanisms.

The final stage focuses on feature delivery efficiency, advocating requirement‑driven testing, improving requirement quality with 5W1H models, and leveraging agile practices to deliver small, test‑validated increments quickly.

Overall, the article encourages teams to evolve testing from a isolated activity to an integral part of the development pipeline, ultimately enhancing product delivery speed and quality.