Why DevTestOps is going to peak in the coming decade

Being a marketer in the tech industry for five years has exposed me to a lot of buzzwords and trends, where some stick while others come and go within the year.

Reflecting on this made me wonder what makes a trend transcend out of the pains of the practitioners’ hearts into the budgets of business leaders ready to put aside thousands of dollars for solutions for it. What key factors are necessary to give impetus to a new industry centered around one buzzword?

One of the terms I’ve seen rise to the top of the list is DevSecOps, the practice of embedding security testing earlier in the development lifecycle. One source notes “The DevSecOps market size is expected to grow from USD 1.5 billion in 2018 to USD 5.9 billion by 2023.” DevSecOps is “hot”, and has become a line item in many IT budgets today - but how did it get here?

A scene from the movie Zoolander of Mugatu and Katinka with the word DevSecOps so hot right now.

In the last decade, huge corporations including Adobe, Yahoo, Uber, Equifax, JP Morgan Chase have had major data breaches. These breaches leaked credit card information, contact information, identities - our valuables. If physical security is important, then with our very livelihoods existing in the cloud, it’s no surprise that digital security and DevSecOps has been such a fast-growing industry.

...But what about DevTestOps?

Last year, we published an e-book about DevTestOps, released an extensive DevTestOps survey on the effects DevOps has on testing and quality, and defined Principles for DevTestOps on testingindevops.org. Why coin another term?

Some can argue that the worst thing that bad quality can cause is poor UX, like the iOS 13 being buggier than ever. #firstworldproblems

But what about the infamous Boeing 737 MAX software bug that caused two fatal crashes? What about the more recent Boeing 737 bug causing cockpit screens to go blank if pilots land on specific runways? What about the software bug in radiation therapy devices that caused them to shoot high-powered x-rays, killing patients? If peoples’ livelihoods need to be secured, peoples’ very lives surely do too.

Okay, so you don’t work in an industry where lives are directly on the line. But even Apple has learned from their mistakes and overhauled the way they test in response to their buggy iOS 13 debut.

But software testing shouldn’t need detrimental quality failures in order to come into the limelight.

On the upside, 2019 was a huge year for software testing. Not only did a sleepy industry meet a revival of sorts as cutting edge tech such as AI/ML and SaaS finally trickled down to software testing vendors, but the testing landscape started shifting as more and more players touted these new features as necessity for success.

It even got to the point where Gartner would declare that 2019 was the last year that they would publish a Magic Quadrant on Software Test Automation based on the standards they had been using. The traditional perspective on software test automation doesn’t cover newer innovative solutions that serve complex, dynamic apps, and of course, DevOps:

“Gartner anticipates that this research will be the final Magic Quadrant for Software Test Automation, as it no longer provides the correct perspective for understanding and evaluating the vendor landscape. The delineation in functionality between various tools — with different focus areas such as web testing, mobile testing, packaged application testing, business intelligence (BI)/warehouse testing, embedded software testing or data testing — and the increasingly tighter integrations in the DevOps toolchain require a new perspective on this market.”

So what now?

We keep cycling through trends but not hitting the root cause of poor quality: not testing in the pipeline. Only then can you shift left and right and ship changes and fixes quickly to production. You can’t jump on it without a strong foundation.

A graph showing cycling through trends but not hitting the root cause of poor quality: not testing in the pipeline.

Testing needs to evolve to keep up, and business people need to invest in testing. Automating the drudgery testing work such as regression testing takes a significant investment. This extends to the “toil” involved in maintaining infrastructure for delivering a product. Automation pays off in the long term with time freed up for the innovation that helps businesses get ahead.

More and more teams are taking a holistic approach to quality, collaborating across roles to test continuously around that infinite DevOps loop. Modern Testing Principles are gaining traction. Testers and QA professionals are moving into coaching and quality advocate roles, helping developers learn testing skills, supporting code with a safety net of automated tests at unit, API/service and UI levels.

These quality coaches and advocates also help with the shift to the left, where teams start testing feature ideas before any code is written. They pare those features down into minimal end to end slices to get quick feedback from customer usage in production.

So what should we do in 2020 and beyond?

Test throughout the delivery pipeline. Democratize testing to it doesn’t become a bottleneck or a burden, but a baked-in process. Put a line item in your budget for DevTestOps solutions - the new age of test automation tools that even the likes of Gartner has even yielded to. You may have heard this Bill Gates’s quote before:

“We always overestimate the change that will occur in the next two years and underestimate the change that will occur in the next ten. Don't let yourself be lulled into inaction.”

So don’t wait for 2020 to be a decade of major software defects. DevTestOps is the next big thing, and you can start taking action now.

> Try mabl: The first test automation solution built for DevOps