AI & automation vs humans: the future of software testing?
AI and automation testing are poised to eventually become a valuable asset for quality assurance (QA) testers and change the way testing enterprises operate.
The reality today is that AI and automation are taking over a large part of businesses practices which are usually conducted manually.
But, if AI and automation are going to lead to new ways in how QA experts operate or how software testing and technology are tested, it might take a while before AI or automation effectively replaces traditional testing procedures led by humans, according to new research from Panaya.
The survey, which was conducted by Vanson Bourne in August 2018 involved 300 professionals from the US and the UK.
Results from Panaya’s survey
Based on the survey, Panaya suggests that the quality assurance is “evolving”, but companies are finding that their efforts to implement change and rapidly speed up the process of testing, benefit more from “humans rather then automating processes,” Rafi Kretchmer, chief marketing officer at Panaya said.
Using Agile methods and DevOps is often viewed as one of the best methods for inspiring change within a team, department, or an entire organisation, but the report states that “Agile methodologies and DevOps has shifted development away from traditional silos, but as the role of QA moves closer to the end user, the definition of testers has changed.”
Their findings suggest that 90% of the companies they asked in their poll said that automation testing allows testers to perform their tests quicker, but achieving quality seems to be a problem among testers. According to their findings, as much as two-thirds (66%) struggle with “merely deciding what to test”.
Survey participants were also asked about how DevOps will impact the process of testing over the next couple of years. According to the poll, 53% of respondents said that testers will take more strident steps to improve quality, which was a 13% increase compared to last year’s survey results.
There has always been significant evidence that adopting Agile practices and values improves the speed of software experts, teams and businesses. Nowadays, many Agile teams are also cross-functional, which means, developers and testers can collaborate much more closely than they previously did in the past.
This has led to an increase in Agile testing in large businesses. Software companies such as Google, Apple, and Amazon are among some of the biggest companies to use Agile practices within their organisations. According to VersionOne’s State of Agile report in 2017, 94% of organisations adopt Agile practices in some form.
When asked about the effects of Agile practices, the vast majority of respondents surveyed reported that implementing Agile processes increases productivity, effort, and efficiency.
The survey results show that 28% of respondents said that Agile methods improve the collaboration between testers, developer, and other team members, and 20% said testing teams are more involved during the early stages of the project’s lifecycle as it helps them identify and rectify any testing issues early-on.
While, productive agile test procedures boost productivity among testers and developer, automation, however – still remains low.
A majority of 61% respondents stated that they automate 50% or less of their functional testing. The survey results also show that 47% of respondents said that they only automate up to 40%, and just under a third (32%) automate up to 30% of tests.
The report also states that deciding what to test and risk-based software testing (RBT) is linked together, as 32% of respondents believe that RBT enhances quality, to help prioritise tests and allocate testing efforts, and 26% view it as a method to identify and minimise risk factors.
Commenting on the benefits of RBT, Kretchmer said: “Risk-based testing offers a variety of key benefits, but by helping solve the issue of what to test, organisations can better allocate test efforts, eliminating wasted time and expenses testing what is out of scope.”
Future of the software testing industry
Regardless of the hype surrounding AI and automation and its role in the future of the software testing industry, Panaya believes that functional testing in today’s organisation is a key indicator to “increase investment” in the skills development of QA professionals.
“The companies that see this are making the investment to get them past the QA bottleneck without sacrificing velocity or quality,” Kretchmer said.
As we transition into a more mechanically-inclined world, it seems as though, based on the finding of the survey, a “human touch” will remain important in the software testing industry.