Easier-to-use tools increase take-up of test automation
Test automation has been around for a long time, yet it has always been a struggle to get the most out of it or even convince project teams to use it. In this blogpost our test expert Magnus Loveman writes about how you increase your return of investment.
The World Quality Report (WQR) points out that more and more practitioners are talking about in-sprint automation, as well as automation in all parts of the QA lifecycle, and not just in execution. The growth in Agile and DevOps methodologies has certainly helped in this respect since they really can't work properly without automation. Beyond that, however, I believe a factor in this trend is the test tools themselves. They have become easier to use, something that many people might say is about time since test automation was first introduced around 30 years ago.
In today’s digital world, the frequency with which applications change continues to grow. So, in 2020 the most important step is surely to automate. And it is clear from the WQR research that most organizations are using test automation, which is great news. However, only 37% of this year’s survey respondents say that they get a Return on Investment from their automation efforts. This is interesting and I suggest that if you are using test automation, you need to ensure you measure its effectiveness and apply it in a wider area than just system testing. For example, automate test data preparation, as well as some of the smaller, repetitive testing tasks.
Adapt the tool to the job
Perhaps one of the barriers to achieving the desired ROI is how automation tools are introduced and monitored. This needs to be done properly and is only possible by implementing the right framework around the tool. The WQR makes a recommendation to opt for automation framework designs that are intuitive and dynamic.
Such frameworks need to address the fact that, while test tools are a lot easier to use than they used to be, they still try to be everything to all projects. But this isn’t required. I would urge avoiding trying to use all the functionality of a tool. You don’t need to force it into your project. Rather, use the automation intuitively and adapt it to the project phase you are in. This is an area where we are gradually seeing the automated assignment of priorities through test management tools, and even test tools automatically creating the test data.
This year’s WQR survey responses reveal that the skillset required for automation remains a tough ask, perhaps because the tools are getting smarter, but the teams are not yet sufficiently skilled to take advantage of them. Often those skills can be limited to the developer community, but with some test automation tools lowering the technical skills threshold today compared with previous years, you have two different types of skillset to consider when you are planning to start using automation. The first is to focus on the technical route (employ developer-minded testers), while the second applies to a code-minimized testing approach, were you don’t need developer skills. The latter can be achieved with specific frameworks, or even codeless tools, which are no longer just a gimmick. Above all, it is crucial to make an informed decision and look at what is needed, where it is needed in your project/organization, and when test automation needs to be there.
A quick word about smart automation
Finally, test tools are continuing to develop, closely following the direction of available technologies. Currently, some tools are venturing into artificial intelligence (AI) and machine learning (ML) to tackle the ultimate technical test automation challenge, which is testing applications that are visually and/or functionally dynamic. While it is important to understand the true capabilities of your test tool(s), including any intelligence or learning capabilities, it is crucial to ask if your project really needs such ‘smart’ test tools at present. These technologies are new to the market with many still very much in development. So, rather than immediately jumping on the bandwagon, perhaps AI and ML are simply worth keeping a watchful eye on over the months and years to come as their value in automated testing evolves.
If you have further questions on this subject please reach out to me.