Test automation is high on the business agenda today.
All, or almost all, companies are working on this aspect with the aim of reducing the effort of performing manual testing, avoiding the rotation of people in projects, motivating employees and gaining time for the commercialization of products and services. Now, before you jump into the automation pool, there are a number of main points that need to be raised to get the project right and the goal we're looking for.
The first thing to put on the table and analyze is that there are three important areas that need to be covered in any test automation process: tools, test environment and scope; although it is not necessary to do so in this order, because all of them are fed back.
In the field of tools, the question we need to ask ourselves is which tool meets our needs. Here the important thing is to analyze what needs we have and observe the types of validation, architecture and tools that allow us to build our automation project, manage and report. It is not the same to select a test automation tool if we are thinking about web or mobile than if we are thinking of an SAP system, because, in this case, we would have to take a more specific approach to technological support and we would have to select the tools that provide us with support for it.
If we move to test environments, what we need is software, licenses, and infrastructure, as well as a run log and parameterization. Therefore, the test data management strategy and the requirements for continuous testing must be defined. In many cases, it's important to note that to get these test environments up and running, you'll need to have a lot of data. This is one aspect we can't ignore.
In terms of scope, what we have to value is what we want to automate. The initial evaluation should aim to define which business processes we are going to automate. To do this, we have to identify what kind of applications we have repetitive tasks in and what we want to automate. Likewise, we need to define at what level we will implement our test automation.
Once we have these three well-established fundamentals, we only have to look at a few additional steps, such as identifying the pilot application, identifying small validation steps, which will have to be performed step by step to properly manage expectations, and automation criteria, which are used to know the stability of the application in a test state. In this sense, we need to be sure what will be used and assess the life cycles of the application; whether it will be used for a long time, if it is obsolete, or if the testing time is acceptable.
Finally, but very importantly we must look at the ROI (Return On Investment), taking into account both the automation costs and the costs that we are going to have. Here the important thing is to estimate the break-even point for what needs to be taken into account the effort involved in development, the support/maintenance that has to be performed and the tasks that we tend to do manually. The balance will be found when we reach approximately 80% profit (when considering the use of automation) and 20% effort (of what would be the total amount to automate 100% of the scope).
In this sense, each organization must define the automation levels that it will implement and its level of scales, starting from 0 that correspond to having no automation and gradually going through levels 1 to 4. The first level corresponds to a basic level; the second with a more productive level in which we already have a roadmap, defined objectives and KPIs; the third with a more consolidated automation framework in which we already work in processes and not just in applications; and a quarter and one more progress in which we have already introduced artificial intelligence. These levels give a priority reference to guide us and ideally have a detailed priority scale. Thus, as we increase coverage by 60-80%, investment begins to decline. This is an important aspect to keep in mind before you start.
In short, it's important to focus on how test automation is going to be used, based on good market practices, how to get the ROI you need and how you're going to measure it.
If you missed the session, you could watch it here
Eduardo Amaral, associate director Quality Manager Noesis-Altia