5 top-tips for tackling continuous testing
One of the biggest challenges facing today’s CIO is the pressure to roll out new IT innovations while managing the increased risk of technical failure. Our dependency on complex IT solutions, from the integration of consumer-facing apps with back-office core systems, alongside cloud, IoT and AI technologies, has significantly broadened the technology portfolio of the modern business.
And, on top of that, we want faster delivery and deployment through agile and DevOps delivery models. Therefore, it’s hardly surprising that many organisations today are struggling to adapt their quality assurance (QA) and testing processes to keep up.
Some software failures are acceptable and can provide a valuable learning opportunity, especially if quick roll-back options are available, but for established enterprises some software failures must be prevented at all costs.
This has made it critical for organisations to include transformation of QA and test approaches in their transition to new delivery models, but where is the best place to start?
Improve requirements definition in agile teams
Recent research reveals that most agile software development teams spend 40-70% of their time clarifying the requirements in order to just understand what should be covered in tests. To overcome this, teams must invest more time in designing relevant test scenarios via Model Based Testing (MBT) and Behaviour Driven Development (BDD) techniques.
This will help them address and resolve any gaps or ambiguities more quickly and provide the opportunity to generate more test cases automatically in the very early stages of each sprint.
Optimise for test tool consistency
Each autonomous project team or squad will use their own choice of test automation tools and test infrastructure provisioning solutions. At the beginning of the process, it is not uncommon that teams spend 30% of their time just to get the test environments right. This causes unacceptable delays, or a risky situation where relevant tests get dropped.
This issue can be reduced with the help of a centralised support Center of Excellence (CoE) for test environment management, test automation tools and the containerisation and virtualisation of environments. With the right technology, such a team can provide project teams with on-demand provisioning of test environments and re-usable test automation frameworks.
Review the approach to test data management
Enterprises rely on multiple, interconnected systems which means the data required for each test is spread across multiple databases. This makes the creation of the test data a highly time-consuming process.
We recommended that organisations define their test data priorities and identify tools that will meet their needs now and in the future, and establish a central test data support CoE.
Across the board, most organisations continue to struggle with automation in testing and the overall approach in the enterprise remains fragmented. Low levels of test automation make it challenging to secure the correct coverage of testing in the shorter sprint-cycles and prevent teams from delivering confident results.
The solution lies in improved test orchestration which combines manual and automated tasks in a holistic fashion, creating an efficient process with right QA coverage.
One of the most common QA dilemmas across both Agile teams and DevOps is ensuring teams have enough guidance and support, without removing their autonomy to be fast and flexible. Each team must be able to work independently, but also integrate successfully to meet the goals and objectives of the larger team structure.
Smart dash-boarding and smarter automated testing technologies are beneficial to help teams increase the speed of deployments in a more complex and highly integrated IT environment.
We can predict with certainty that the IT landscape will continue to evolve over the next twelve months, and with it comes another certainty that organisations will be required to test more frequently, across more complex and integrated solutions, and in shorter timeframes.
However, those that make positive steps now are best placed to increase quality, reduce cycle times and use resources more efficiently.
By Mark Buenen, global leader, digital assurance and testing, Capgemini Group