2 min Devops

AI saves development time, but inefficiencies still cause losses

AI saves development time, but inefficiencies still cause losses

A new Atlassian study reveals a paradox. Developers may save up to 10 hours per week with AI tools, but lose just as much time due to organizational inefficiencies.

Almost all developers (99 percent) report time savings from AI tools, with 68 percent saving more than 10 hours per week, according to Atlassian’s 2025 DevEx survey of 3,500 developers and managers. These gains come not only from coding tasks but especially from non-coding activities. Developers use the time freed up for code improvements, new functionalities, and documentation.

The survey reveals an alarming pattern: 63 percent of developers say their managers don’t understand their problems, up from 44 percent last year. This empathy gap arises because managers credit AI with saving time without addressing existing points of frustration. As a result, they expect faster delivery, while the underlying problems remain.

Balancing time savings and frustration

In 2024, developers saw little productivity gains from AI tools. That has changed dramatically. At the same time, the picture remains problematic. Developers report that they spend 50 percent of their time on non-coding tasks, with 90 percent losing 6 hours or more per week to organizational inefficiencies.

“We’re right back where we started,” the study states. “Developers saving 10 hours a week using AI and losing 10 hours a week to inefficiencies.”

The biggest time wasters are searching for information (services, docs, APIs), adapting to new technology, and switching between tools. It is striking that technical debt disappeared from the top 5, while collaboration with other teams has become a bigger bottleneck.

Systematic approach necessary

Atlassian emphasizes that AI is not a silver bullet. The data shows that AI can improve the developer experience, but only if it is used to solve real pain points.

The first step is always the same: talk to your developers. Organizations need to gain a deep understanding of the points of frustration and test solutions together with developers. Sometimes AI is part of the answer, sometimes it’s simple things like self-service materials.

Developers also have a role to play. They need to help managers become aware of challenges by framing them in terms of impact. This makes it easier for leaders to set priorities.

Tip: Atlassian’s System of Work vision is taking shape with Teamwork Collection and Rovo AI Agents