2019 is an exciting year for us @TestResults.io. We are building new features to our tool chain which will allow us to deliver automated test cases to our customers even faster than now. The world doesn’t stop at our service though. We are also monitoring the current trends in Test Automation to always be one step ahead. Here is a glance of the trends we believe will be popular in the market in 2019. We don’t necessary always agree with them, but it’s good to have an overview of what is going on in the test automation world.
You might have heard this fancy term but what does it mean? The simplest explanation would be to “move” the testing phase closer to the development phase (if you look at the software development process in a linear way). However, it is just a too vivid way to explain this trend. Shift left should not be considered as moving the testing earlier so that testing can be avoided later. It should rather be considered as making sure that whatever can be tested in the earliest possible phase will be tested, but only if a corrective development actions will also take place. It’s not just a matter of shifting the testing but adjusting the whole development process so that tests can be performed and reacted upon. You should think of it as a way to identify risks and problems earlier than now, but not to forget that it is a constant process which should not be omitted at later stages.
What does it mean in terms of Test Automation? Basically it all depends on how your development process will be adjusted e.g. you will not be able to create more UI tests if this will not be supported by delivering the UI earlier.
You have probably heard of the new fancy trend in software development called Low-Code. How does it work? Have you ever built a WordPress page with some additional plugins and forms? In a very big simplification it works in similar way. The developers are building their software by configuring and combining some functions into workflows and pages. Basically they are not coding each module from scratch but instead are just customizing them. All of those modules will in the end build their software. How does it effect a tester and test automation though? You might hear a theory that you will be able to drop some automated unit tests since each single module will be already unit tested prior to its usage. In this case you should focus your attention on building as many integration tests as possible. After all, all of the modules need to work flawlessly together.
Everybody speaks about Artificial Intelligence in test automation but the bottom line is: AI as we think about it doesn’t exist. Period. The trend we are seeing now should be rather called: a smarter way to identify controls. How many times your automated test case failed due to the object not being found? Lets say the automation id has changed, or the text label you have used before has been modified. This trend addresses this problem and allows you to have a more robust way to find elements. If you thought that “AI testing” will magically test for you, then fortunately for us you were wrong. Otherwise, we the testers, would become obsolete.
The more you know the more valuable employee you are. The same principal applies for Data Analysis. If we know more about the tested system we can determine better were the pitfalls are. The analysis of (meaningful) data allow you to check where the biggest problems in your software are occurring or might occur. Imagine that based on a smart algorithms you can determine where you can expect the biggest density of your issues and therefore understand what should the biggest focus in your development be. It doesn’t stop there. You also start analyzing data of your test automation runs, to understand where your test automation solution improvements should be made. Data Analysis is a powerful way to optimize both your software and test approach in a preventive way.
Application monitoring is all about continuously checking of your software for negative impacts of the newly introduced features/fixes. Not only does the application monitoring allows you to check response times, traffic, errors, but also run the functional tests all the time. You might ask yourself how is this approach different than your current continuous integration approach or just a normal regression testing? The answer is quite simple. Application monitoring is performed on the life and deployed system and not in your isolated test environment. The tests are running 24 hours a day, 7 days per week and are a great indicator of the current stability of your productive system. Application monitoring allows you to minimize the risk of your customers finding problems in your system before you do. Naturally, it will not allow you to magically heal your product, but at least you will be one step ahead and you will be alerted about the things which went wrong.
Not each single trend in 2019 is brand new and not each single trend needs to be immediately followed. You need to find out what is applicable to your software development landscape. Your other option is not to follow any of those and leave it to the TestResults.io experts. Just click below :).