Software testing remains to be a struggle for development teams around the world, according to an annual global survey conducted by GitLab.
Nearly 4,300 respondents to the survey, including developers, engineers, members of management, and some C-level executives in various industries, all identified software testing as the most likely culprit of slowing down software development.
“For the third year in a row a majority of survey takers resoundingly pointed to testing as the area most likely to cause delays,” GitLab stated in its survey.
Of respondents who completed the survey from February to March 2021, 25% reported that their teams had no test automation in place. Another 9% surveyed admitted that their teams haven’t shifted testing far enough left to where they thought testing needed to be.
On a brighter note, 25% of the respondents reported working in an environment that included full-test automation. That number was an improvement upon previous year’s results, which found only roughly half that reporting full testing capabilities.
Almost 28% of respondents to this year’s survey reported partial progress in implementing automated software testing.
“Roughly 34% of survey takers said developers test some of their own code (up from 31% last year) and 32% said automated testing happens as code is written, a big jump from 25% in 2020,” according to GitLab.
Even with the progress, anecdotal responses to the survey provided more sharp and direct criticisms of software testing, its impact on delivery, and an overall frustration with the process.
“Testing delays everything,” one respondent stated.
Other responses to the survey listed poor communication between teams, resources for and complexity of code reviews, and poor planning as more negative factors that hinder testing.
On the subject of communication, the interaction between teams played a large role in holding back testing, according to anecdotal responses.
“Communication between business and development is hard,” one respondent stated.
“A majority of our delays are created by multiple teams being involved, under different management umbrellas, without one cohesive vision at the top layer.”
On the code review front, respondents voiced struggles around not only finding the resources to conduct the software evaluations, but also following them through to completion. And when code reviews did get underway, the process would drag on and on.
“Finding someone for code review can be hard (1 day average). After that business tests take time to be complete (2-4 days on average).”
And that’s if the code gets reviewed at all. Lack of communication overall, coupled with struggles with code review, can have companies making tough decisions.
“Developers are sometimes unaware they have to do code reviews. They aren't sure how to perform them and if they are effective. Sometimes they are skipped so the process can go through,” another respondent stated.
It’s worth noting the makeup of the audience that completed the survey because it gives a window into who’s telling the story around the struggle with testing.
Nearly half (41%) of respondents were either software developers or software engineers, making them the largest representative group to complete the survey. Those in leadership positions represented only about 14% of the total respondents. Only 3% of those who answered were members of the technology c-suite, whether CIO, CTO, or Vice President-level executives.
That means the message around struggling with testing isn’t something that’s coming from management or leadership. It’s a story that the people on the frontlines -- those closest to the actual creation of the software -- are telling us.
Is testing a struggle for your team? Getting started with automated software testing can be tough but it's possible with an experienced team. We help companies build better, safer, higher-quality technology everyday, cutting delivery times significantly as the process improves.