Atlassian says AI has created an 'unexpected paradox' for software developers – they're saving over 10 hours a week, but they’re still overworked and losing an equal amount of time due to ‘organizational inefficiencies’
While developers are flocking to AI tools, productivity gains are being offset by other tedious manual tasks and poor leadership
Software developers are saving more than a full day’s work each week through AI tools, new research shows, but they’re losing time in other critical areas.
Atlassian’s State of DevEx report found that AI adoption among developers has surged over the last year, helping to drive marked productivity improvements and reduce manual toil.
More than two-thirds (68%) of developers told Atlassian they’re saving at least 10 hours per week using AI tools, which marks a significant increase compared to last year when just 46% of devs reported productivity gains.
These time savings are being put to good use, the study noted, with developers using spare time to focus on improving code quality or building new features.
But while developers are saving time in some areas, pressure is mounting in other areas of their working week. Atlassian said its study uncovered an “unexpected paradox” whereby despite the increased use of AI, overall productivity isn’t keeping pace.
Indeed, 50% of developers said they still lose over 10 hours a week and 90% lose at least six hours or more due to “organizational inefficiencies”.
Common issues identified by both developers and managers include poor communication and collaboration between teams as well as a “lack of clear direction” on projects.
Sign up today and you will receive a free copy of our Future Focus 2025 report - the leading guidance on AI, cybersecurity and other IT challenges as per 700+ senior executives
Another crucial hurdle for developers includes simply finding information, Atlassian noted. This emerged as the leading cause of wasted time for development teams alongside tackling technical debt.
This isn’t the first time Atlassian has highlighted the performance impact of knowledge retrieval. A study from the company in March this year found workers were wasting a quarter of their working week tracking down information relevant to their job or individual tasks.
UK workers in particular wasted an average of nine hours per week, accounting for hundreds of hours annually. The State of DevEx report noted that, for a company with 500 developers, this wasted time equates to a loss of $7.9 million (£5.8m) annually.
Atlassian CTO Rajeev Rajan said that while the report showcases the benefits of AI, it also highlights the need for more efficient cross-departmental processes.
“This pressure-cooked mix of innovation and strain demands a closer look at how AI is reshaping the developer experience, and what that means for the future of software development across the industry,” he said.
“In a year of economic turbulence, rapid AI breakthroughs, and growing expectations to ship faster, developer experience continues to take center stage in leadership conversations”.
Developer challenges need to be acknowledged
Crucially, Atlassian’s research warns of a growing disconnect between management and developers, with the former often lacking a clear understanding of the challenges faced by devs.
Nearly two-thirds (63%) of developers said they feel leaders at their organization don’t understand key pain points faced by teams in their daily activities. This marks a 19% increase compared to last year’s survey.
This is where the question of whether or not AI tools are truly helping developers comes into the equation, Atlassian noted. These tools primarily support developers with coding tasks, which only make up around 16% of their working week - the remainder of their time is spent on general administrative tasks.
“To truly improve the developer experience, the 84% of the workweek spent on less enjoyable tasks needs to be addressed,” the company said.
Addressing these hurdles will be a challenge for C-suites and senior developers alike, the study warned, but a good starting point could be to improve documentation and information sourcing given it was identified as a leading pain point.
Notably, the firm said developers are crying out for the ability to “build and maintain self-serve resources” that they can draw upon with ease. Unlocking efficiency gains in this area alone will reduce manual toil and help streamline processes.
“When developers can self-serve information, they are 4.4 times more productive,” the company said. “To bridge the developer-leadership gap, teams need to communicate challenges early with clear examples and context, identifying friction points.”
Make sure to follow ITPro on Google News to keep tabs on all our latest news, analysis, and reviews.
MORE FROM ITPRO

Ross Kelly is ITPro's News & Analysis Editor, responsible for leading the brand's news output and in-depth reporting on the latest stories from across the business technology landscape. Ross was previously a Staff Writer, during which time he developed a keen interest in cyber security, business leadership, and emerging technologies.
He graduated from Edinburgh Napier University in 2016 with a BA (Hons) in Journalism, and joined ITPro in 2022 after four years working in technology conference research.
For news pitches, you can contact Ross at ross.kelly@futurenet.com, or on Twitter and LinkedIn.
-
AI might help speed up software development, but 81% of devs now spend more time reviewing code – and it’s creating an ‘invisible work’ trend that’s pushing teams to the limitNews While AI is improving productivity and efficiency, many developers are caught up in a vicious cycle of code reviews and bug hunting
-
AWS CEO Matt Garman is bullish on the future of SaaS — Amazon Quick shows there’s a ‘great business opportunity’ with AI-powered softwareNews Matt Garman said fears over the ‘SaaSpocalypse’ were overblown in February, now AWS is making big moves in the SaaS space
-
AI is coming to Ubuntu: Canonical exec teases future AI features and agentic workflow capabilities for version 26.10 — but on a ‘strictly opt-in basis’News A range of new AI features are coming to Ubuntu over the next year, according to maintainers, but only providing they’re of “sufficient maturity and quality”.
-
Everything you need to know about the GitHub Copilot pricing changesNews GitHub Copilot pricing changes mean users will be charged based on consumption, rather than a set number of credits
-
Developers are slacking on AI-generated code safety – here's why it could come back to haunt themNews While organizations are aware of the risks, many are spending little time or effort on tracking artifact versions, origins, and security attestations
-
Four things you need to know about GitHub's AI model training policy – including how to opt outNews Users of certain GitHub Copilot plans will have interaction data used to train AI models, but can opt out
-
'AI doesn't solve the burnout problem. If anything, it amplifies it': AI coding tools might supercharge software development, but working at 'machine speed' has a big impact on developersNews Developers using AI coding tools are shipping products faster, but velocity is creating cracks across the delivery pipeline
-
‘I hope there's a world where AI is is complementary to humans’: Workday CEO vows to support HR workers as Sana integration automates more processes than ever beforeNews Sana from Workday seeks to bring agentic AI to Workday’s systems and beyond with natural language input and third-party connectors
