Do Low-Code Testing Solutions Offer Promise?

Posted Aug 18, 2020

Multiple screens of code

Are you excited over the prospect of more automation?  I’m amped over it, jacked...you could say I’m even a little hyper over it <rolls eyes>.  Hyperautomation is quickly becoming one of the IT buzzwords of the year.  Gartner defines Hyperautomation as “the application of advanced technologies, including AI and ML, to increasingly automate processes and augment humans.” Gartner goes on to say that “Hyperautomation is an unavoidable market state in which organizations must rapidly identify and automate all possible business processes.” I’m all for bionic eyes and a super-powered exoskeleton, but how will Hyperautomation play into application development and test automation?

For starters, Hyperautomation is simply a term Gartner has dubbed for what is essentially the automation of things.  In this case, it’s implied that nearly everything technologically connected needs to be automated to some degree.  This is where the term “low-code” comes into play.  As automation efforts expand to account for software design, modeling, monitoring, analytics, robotic process automation (RPA), business process automation (BPA), and more, we are simultaneously seeing the democratization of these technologies such that any stakeholder can participate in the effort.  This is good news! It means the SDLC will harness automation efforts from everyone involved, not just developers and engineers.  Low-coding allows practitioners to create and test application software with an approach that does not involve computer programming.  It represents a leap beyond record and playback tools that have several inherent limitations, including; scalability, flakiness (which in turn requires coding to maintain robustness), hardcoded data, editing with code, and in general, requiring code for a successful implementation.

With the advent of several low-code test solutions emerging in the marketplace; Sofy.ai, AccelQ, Testim.io, and even ProdPerfect, we’ll begin to see more adoption along the way with more testimonials over their efficacy.  Low-code solutions are not going away and will only become more prevalent as the industry makes continued progress with ML and AI.  Moreover, application development is conducted at a pace that is more expeditious than ever before. The SDLC for applications has been quickened (thanks to increasing adoption of DevOps methodologies and an incredible surge in mobile apps), putting great pressure on conventional automation frameworks to be flakeless and quick to debug, in order to keep up with the rapid pace of development.  Enter low-code solutions to help augment both the development and testing process where speed is paramount. 

Low-code test solutions are like any other tool in the proverbial “toolbox of test automation software”—another tool.  It’s new and shiny and definitely makes things easier, but most likely will not replace your other test tools.  Conventional test automation frameworks and strategies are very relevant and continue to expand and grow as new technologies emerge and time-to-market decreases.  MarketsAndMarkets predicts that globally,  the “Automation Testing Market” is poised to be worth $28.8 Billion by 2024 (was $12.6 Billion in 2019), and grow at a compound annual growth rate of 18%. The report goes on to indicate that functional testing tools will have a larger market size. Low-code test solutions are perfect for functional testing; functional testing is based on business and customer requirements (product teams can now help write tests!) as well as being dispositioned as opaque testing (no knowledge of the inner workings of the application and how the code is assembled).  These attributes are primed for low-code test execution, allowing more time for those who can code to focus on the components that need it.  Consider current CI/CD automation testing, you probably don’t want the product team messing around in your SDE, or for that matter, checking in code (after all, this is low-code). 

Of course, low-code testing is not without its limitations to consider. Firstly, when building test automation to scale, it's important to have systems in place that allow for easy maintenance and sustainability of tests as the suite grows. Low-code platforms allow more personas to embrace testing, leading to inherent challenges in ensuring that the suite is optimized to accelerate software delivery, and not create bottlenecks due to dependencies, flaky tests, etc. Additionally, there are far too many testing scenarios that must be covered and low-code testing will not account for it all.  Security testing, performance testing, user experience testing (usability), compatibility testing, and compliance testing all represent different types of non-functional test methods that are most likely already automated (to some extent), if for no other reason than to reduce time. 

Low-code test solutions have definitely left the station and are moving like a freight train.  While they won’t be a substitute for conventional automation frameworks (at least not yet), they hold serious automation promise through visual test creation, including more people in the SDLC, and elevating everyone to be a practitioner despite their skill level.

Written by

Darryl Jenkins