Almost half of the developers spend 10-25% of the time to correct errors in the finished product

According to a new study, 43% of enterprise developers spend from 10 to 25% of their time debugging and correcting errors in applications at the operational stage.

ClusterHQ analysts came to this conclusion based on a developer survey: a quarter of respondents reported that they had to correct production bugs several times a week.
')
The authors of the study believe that such a waste of time is too great: developers need to focus more on the implementation of new features for users, rather than on the constant correction of errors.

Developers are asked to name the most common problems that they have to solve. 33% noted the inability to recreate the real environment during testing. 27% recalled the interdependence of the product and external systems, making it difficult to integrate testing. 26% pointed out the problem of modeling realistic data needed to test the application.

Researchers also asked respondents to evaluate the priority of bugs in terms of their criticality. 62% of respondents answered that the price of an error missed at the production stage is the highest. 18% consider the most "expensive" stage of development, 7% noted the QA stage, and 6% - testing.

“Our application testing study shows that practices such as testing based on a limited subset of specially generated data are no longer suitable for teams focused on maximizing the time needed to develop valuable features for users,” said Mark Davies, CEO of ClusterHQ. “Progressive software developers understand that in [timely] supplying customers with innovations and refinements, it is necessary to effectively manage the software life cycle along with various levels of infrastructure. This process begins with identifying and eliminating bugs as early as possible so that teams can focus on adding the [features] necessary for the end user. ”

The authors of the study believe that changes in software development and testing will positively affect the company's ability to scale. Another thing is that not all companies want and can make these changes.

88% of respondents would like to be able to test applications on real data during development. However, while on the way to this there are several obstacles. The main one is to ensure the relevance of the test data. So say 23% of respondents. 19.5% consider the main problem is the complexity of simultaneously updating test sets wherever they are stored. The same number of interlocutors mention the difficulty of supporting multiple versions of data. 18% note the problem of controlling access to information, 14% consider copying data from a working system to be an overly laborious process. 6% are sure that the cost of storing huge amounts of excess information will be unreasonably high.

A total of 386 IT professionals were polled. The full research report is available here .