A survey of 300 mobile engineers in the U.S and the United Kingdom (UK) working for organizations with 500 to 10,000 employees finds, on average, respondents are spending five hours per release on low-value tasks involving manual steps, coordination issues and approval bottlenecks, translating into 130 wasted engineering hours annually per developer.

Published by Runway, a provider of a software-as-a-service (SaaS) platform for managing mobile application releases, the survey finds more than three-quarters (77%) said their teams experience incidents requiring delays or hotfixes every three to five releases. Most teams (86%) spend a significant amount of time firefighting instead of building features, the survey also finds.

Despite those issues, however, most respondents said it was either very unlikely (9%) or only somewhat likely that their current release process could lead to missed deadlines, delayed features or negative user impact. Only 17% believe it is somewhat likely that inefficiencies in their current release process would lead to such issues.

Just over half (51%) also view their processes as being “somewhat” efficient, even though 52% also report spending about a third of each release cycle on non-productive or low-value tasks.

Only 11% report that their process is very efficient. Two-thirds also agree that improving centralized release coordination would have a significant (19%) or at least moderate (47%) impact on their processes.

Runway CEO Gabriel Savit said the survey makes it clear that as the overall pace of mobile application development accelerates, the teams building these applications are not becoming more efficient, resulting in higher costs. The challenge is that awareness of the issue is limited because of a general lack of observability and automation, he added.

Much of that issue can be traced back to the tools being employed by mobile application development teams, noted Savit. For example, most teams are relying on general-purpose tools such as project management applications (81%) and communication platforms (68%) for release coordination, including Jira/Asana software (81%), Slack (68%), continuous integration/continuous delivery (CI/CD) tools (53%) and spreadsheets (47%).

Not surprisingly, manual steps (49%) top the list of frustrations, followed by context switching (42%), and cross-functional coordination (41%). Respondents estimated that better communications, processes and collaboration across releases could have prevented 63% of uncaught incident issues.

It’s not clear to what degree these issues are more challenging to manage when building mobile applications but end user tolerance for any type of mobile application quality issue is notoriously low. It’s simply too easy for them to switch to another application. As more individuals participate in the development of these applications in the age of artificial intelligence (AI), it’s only a matter of time before the number of applications being simultaneously built is likely to overwhelm existing processes, making it even more critical to prevent quality issues, noted Savit. The issue at hand now is determining to what degree to revamp those processes now versus waiting for the inevitable crisis, he added.

Ultimately, each organization is going to have to revisit how they build and deploy applications in the AI era. The only thing left to determine is the how and when, as the pace of application development continues to steadily increase beyond what any workflow designed for a previous era of software development can effectively support.