AI adoption is saving builders a big period of time, however a number of friction factors nonetheless exist all through the software program growth life cycle that builders are nonetheless dropping time to.
In response to Atlassian’s 2025 State of Developer Expertise report, 99% of builders are saving time by utilizing AI, and 68% save a minimum of 10 hours every week. For the survey, Atlassian gathered responses from 3,500 builders from the US, Australia, France, Germany, the UK, and India in March of this yr.
This yr’s outcomes are an enormous change from final yr, when a majority of survey respondents had but to see any productiveness features.
Of the builders saving time with AI, the highest 4 methods they’re reallocating their time embody enhancing code high quality, growing new options, enhancing engineering tradition, and growing documentation.
Jellyfish additionally not too long ago launched its 2025 State of Engineering Administration report, the place it discovered that 90% of growth groups are utilizing AI coding instruments, and 62% consider they’re seeing a minimum of a 25% improve in developer velocity due to these instruments.
Jellyfish’s co-founder and CEO Andrew Lau stated that the productiveness features detailed within the report haven’t been in step with the 10x productiveness will increase that had been predicted.
“AI adoption in software program engineering is not non-compulsory, however unlocking its full worth calls for greater than entry,” he stated. “It requires intentional measurement, structured enablement, and cultural funding. This isn’t a tooling improve – it’s an organizational transformation.”
Atlassian additionally famous that builders solely spend 16% of their time coding, and far of the present funding in AI is in AI coding assistants. “Coding isn’t a friction level for builders, which is why coding assistants can improve the expertise with out enhancing it,” the corporate wrote in a weblog put up.
This yr’s survey discovered {that a} vital period of time is being misplaced to non-coding duties. Fifty p.c of respondents reported dropping over 10 hours per week, 40% lose six to 10 hours, and 10% lose lower than 5 hours.
“So we’re proper again the place we began, with builders saving 10 hours every week utilizing AI and dropping 10 hours every week to inefficiencies. Enhancing the developer expertise requires a scientific method to understanding and resolving developer friction factors,” Atlassian wrote.
In response to the survey outcomes, the highest time wasters are discovering info, adopting new know-how, and context switching between instruments.
Making issues worse, 63% of respondents consider that leaders don’t perceive their ache factors, a rise from final yr when solely 44% felt the identical. Atlassian believes this can be attributable to leaders accepting the time financial savings achieved by AI with out addressing underlying points.
Atlassian recommends that leaders use knowledge to uncover obstacles, and take heed to engineering groups intently about their considerations surrounding roadblocks, unclear necessities, or heavy workloads.
“Builders are closest to the work and sometimes see issues first. However these insights solely assist in the event that they’re clearly communicated – with particular examples, measurable affect, and enterprise context. Leaders want to ask and act on this enter by retrospectives, surveys, and skip-level conferences that foster open dialogue,” the corporate wrote within the report.
