New voke study finds service virtualization saves time and resources — and results in a better product

Although more companies are adopting service virtualization, the use of automation and simulation in software testing has yet to achieve proverbial critical mass.

But after leafing through a new report on the returns on investment that service virtualization creates, I was left with a question: How can a business that produces software justify not investing in this technology?

The research report, prepared by Minden, Nev.-based voke media, was presented recently in a webinar (click here to watch the replay) by the company’s founder, Theresa Lanowitz, along with Scott Edwards, director of product marketing of CA Technologies, a sponsor of this blog.

In a survey between August 2014 and October 2014, voke quizzed 505 people in both technology and non-tech companies of varying sizes about their use of service virtualization, along with the results they experienced.

There was an awful lot in the webinar to unpack. That’s why I urge you to watch the whole thing yourself. For purposes of the blog, I’ll go over the results in two blog posts. Next week, I’ll discuss voke’s findings on the challenges that respondents found in getting their organizations to adopt the testing and simulation technology, along with a few suggestions that Lanowitz and Edwards had for getting over those humps.

A few highlights of the ROI that service virtualization generated for the respondents in the voke study:

  • Reduced wait times. On average, survey respondents had to wait an average of 32 days to get access to tech systems they needed to do their jobs, such as datasets, user interfaces, applications and components. After implementing service virtualization, the average wait time shrank to one day, with a median of wait time of only one hour.
  • Less time to reproduce defects. Another issue for participants was providing the test and the environment containing a software defect. The study found 38 percent of participants had cut their defect reproduction time by 50 percent or more.
  • Fewer production defects. Overall, 36 percent of participants reported lowering production defects by more than 41 percent.
  • Cutting total defects. In a similar vein, 46 percent of participants saw a reduction of more than 41 percent in total defects. 

More test productivity

Overall, voke found, participants in the study said service virtualization was a way to enable more test automation, test earlier, reduce testing cycles, and test what is required. For instance:

  • Increased test coverage. Some 20 percent of participants more than doubled their test coverage after using service virtualization, voke found.
  • Better test execution. And some 26 percent of respondents doubled their test execution rates – or better.
  • Reduced test cycle time. Voke fond 34 percent of participants experienced a 50 percent or greater reduction in test cycle times.
  • Shorter cycle times to software release. A total of 40 percent of participants saw software release cycles drop by 40 percent or more.

Summary

For organizations of every size and stripe, software is part of what they produce now. Putting out software with defects is a recipe for turning your customers off –- and sending them to the competition.

“You want to protect your brand,” Lanowitz said. “Software is part of your brand promise. Virtualize the (software development) life cycle as much as you can.”

Employing service virtualization results in cutting wait times for developers and testers alike, she noted. “Use service virtualization to get quick, measurable returns on investment on reducing defect production times, the total number of defects” and software release cycles, among other things, she said.

Edwards added that when organizations implement service virtualization, they “view it as a way to ignite innovation.”

“The practices you used in the past must change to be able to compete in the application economy,” he said.

Some 94 percent of chief information officers feel pressure to deliver software faster than they had before, Edwards added.

But that effort of producing software faster will be for naught if defects crop up in the finished product, he said. “It all goes down the drain if the quality is bad and not on part with what the customer expects.”