Capgemini and Sogeti interviewed thousands of technologists and technology professionals for their annual World Quality Report to understand about the state of software quality practice. The World Quality Report 2021-22, sponsored by Micro Focus, is Capgemini and Sogeti’s 13th such endeavor, providing the writers with more than a decade of data to highlight not only the condition of software quality, but also how trends—and the culture of quality—have changed.

The study is highly extensive, with 72 pages, and what this year’s report reveals may surprise you. You’ll want to read the entire report for all the information, but here are the top eight takeaways from the most recent World Quality Report—everything you need to know, plus comments from the authors and other experts who contributed their perspectives on the findings.

1. Testing and quality are playing a bigger role

The report’s first figure depicts QA’s goals: find problems early, move faster, and contribute to business growth or customer pleasure. The figures range from 58 to 62 percent, and they haven’t moved much during the last four years.

To put it another way, almost everyone wants everything. As a consultant, I’d interpret this as immaturity, but the writers of the report view it differently: management is now realizing that quality assurance is about more than discovering flaws. Testing has been pushed beyond a single person, onto everyone on the software delivery team, which may explain the minor decline in QA’s function as “custodian of quality.”

According to Andrew Fullen, head of innovation and technology for Sogeti in the United Kingdom, this means quality will play a larger role and be seen differently.

2. The rise of AI in quality is extending what testers can accomplish

Industry leaders and consultants have been whining about the AI hype for years. However, according to the paper, AI is outsourcing some cognitive work from humans to computers. “These findings highlight the importance of using AI-powered continuous quality and test automation tools to advance modernization initiatives,” said Rohit De Souza, Senior Vice President and General Manager of Micro Focus’ ITOM and ADM product groups, as well as Leader of the CTO Office and Product Security.

“When it comes to AI-powered testing, there are a few important factors to consider, according to Micro Focus chief technologist Don Jackson. “The first, and most important, is the use of computer vision—getting a computer to see objects in the same way that a human sees.” In the past, testers had to rely on the properties exposed by the operating system, and those values would change for the same object every time they switched platforms. Depending on the platform you were using, you could have up to four separate scripts for the same business process. “”Computer vision works by analyzing the visual features of objects in the same way that a human does,” he explained.

Because of the mix, tests are less fragile because they don’t need to employ precise locator strings or classic code (Java, C#, Ruby). Risk-based testing can be done with usage data thanks to log query and analysis tools. None of these are human-replacement tactics. Instead, they’re methods for improving human performance.

3. Testing has room to improve—and is actually improving

The following graph assesses how well the above-mentioned goals were met. Only half of the people polled said that testing was done to save time and money. This suggests that there is still a lot of room for improvement. Last year, the same question received only 37% of the vote. That’s a substantial improvement in just a year, and it could be attributed to how organizations handled COVID-19.

4. COVID-19 response has been integrated into the world

The impact of COVID-19 on software quality was covered in depth in last year’s report. Financial services and manufacturing, for example, have been sluggish to adopt cloud computing, making the transition to remote work more challenging. Companies were more open to out-of-the-box techniques as a result of the pandemic’s time, resource, and availability constraints—not simply working remotely, but also crowd-sourced testing.

As more personnel were required to conduct testing remotely, a need for data-masking services to transform and anonymize production data arose. After a year, the Covid issue had “faded into the woodwork” of software development, and most organizations had incorporated the adjustments into their software delivery processes, according to Jackson. Working remotely, which was previously an option, has now become a standard component of the job. As a result, rather of having a separate part for pandemic replies, the answers to COVID-related questions were incorporated into other sections of the report.

Many Silicon Valley companies have switched to remote-only testing in the last year. This is a significant shift, allowing qualified individuals from all across the country to compete for jobs at Apple, Google, Facebook, and other now-remote-first businesses.

This will restrict the supply of high-end local test- and quality-related jobs, particularly those that require staff to come into the office. In the coming year, companies that refuse to enable remote work or insist on a back-to-the-office policy will experience difficulties in recruiting, retention, and morale. Many employees have become accustomed to working from home and enjoy it, or have relocated to locations with lower living costs and improved quality of life.

5. The lines have blurred between testing, quality, agile, DevOps, and digital transformation

The amount of overlap between these themes in prior World Quality Reports, as well as how they have converged, is notable. Instead, then having separate surveys for “Agile Testing,” “Digital Transformation,” “Testing in DevOps,” and so on, this year’s study integrated them all into one.

While there was a section on digital transformation in the most current WQR, it is becoming evident that agile, DevOps, and quality are all requirements, if not the mechanism, for digital transformation.

6. Test tooling and automation are becoming the norm

Test automation ranked last among the improvement priorities, with only 46% of respondents identifying it as a top priority. The temptation is to interpret this as “not essential,” but it could also signify that corporations are successfully testing. In some circumstances, test tooling has progressed from a strategic endeavor to an everyday occurrence. “I’d much rather have a tiny percentage of good-quality automation that makes a difference than automating everything,” Fullen of Sogeti said.

Companies are finally starting to grasp the necessity to hire for and build software development skills in test, according to Shivakurma Balasubramaniyan, vice president of financial services, data assurance and quality at Capgemini and another co-author of the research.

That means they don’t have an “automation initiative” when they notice a flaw. They refer to it as hiring, a job description, restructuring, or training instead.

7. There’s a growing recognition that testing is actually a skill set

Balasubramaniyan’s remark leads to the team’s most popular success factor: having the appropriate skills. This includes AI, DevOps, continuous integration, and other things, but most crucially, actual testing and test design.

The biggest problem respondents experienced, according to Kirthy Chennian, vice president for North America at Capgemini and a report co-author, was gaining sufficient agile testing skills on software teams. That might sound strange at first, given how long it had been since many respondents claimed they had completed an agile change.

This is because the person executing the task has shifted from a specialist role to everyone involved in software delivery, according to Chennian. Organizations are recognizing that testing is a repeatable skill that can be developed, trained, and hired for after a few years of assuming it was simple and anyone could do it.

8. The use of Docker/Kubernetes for testing and test environments is accelerating

Companies that use tools to test need servers to run against the most recent build. Just over half of the firms polled have on-demand, fit-for-purpose test environments, and the Docker image, likely running in a Kubernetes cluster, appears to be the de facto standard. This makes testing very simple for businesses, especially because Kubernetes is supported by every major cloud provider.

What’s next in QA

Nearly half of the firms polled do not have on-demand test environments, which indicates that more than half do. The key questions are what matters to your company and where it stands in comparison to its competitors. The latest World Quality Report could be a wake-up call if you don’t have an on-demand testing environment.

These major takeaways are broad in scope and do not include specific industries like automotive, energy, financial services, healthcare, government, high tech, or telecommunications. However, the report’s more thorough breakdowns devote 20 pages to data particular to each industry. To learn more, download the full World Quality Report 2021-22.

For more info:

Also Read: