Despite widespread excitement about AI’s impact on software development, many engineering teams experience underwhelming productivity gains. Atlassian research highlights that real value emerges when AI addresses organizational friction around code, not just the coding itself.
- AI time savings limited if focused only on code generation
- Most friction lies in collaboration and information management
- Scaling AI across the ‘system of work’ boosts team productivity
What happened
Atlassian’s State of Developer Experience 2025 report and commentary from their CTO, Andrew Boyagi, reveal a nuanced reality for AI in software engineering. While AI tools help developers code faster, the overall productivity improvements remain muted because the bulk of team friction is outside direct coding tasks. These include planning, decision-making, documentation, and communication challenges that traditional AI assistance in code editors does not address effectively.
Boyagi shares real-world examples from Atlassian’s own multi-national teams that leverage integrated collaboration platforms such as Jira, Confluence, Loom, and Rovo, combined with AI-driven summarization and task extraction, to reduce the need for time-consuming meetings. This approach enhances asynchronous workflows and helps keep distributed teams aligned without increasing communication noise.
Why it matters
Engineering organizations face an often overlooked productivity problem: nearly 25% of knowledge workers’ time is spent searching for information, and many developers lose more than 10 hours weekly navigating documentation and references. Simply accelerating code typing does not address these inefficiencies. AI’s potential lies in simplifying the entire software delivery ecosystem—from collaboration and knowledge management to decision transparency.
Moreover, deploying AI in an ad hoc fashion within siloed tools can inadvertently increase cognitive overload and duplicate communication channels. Atlassian’s perspective emphasizes designing AI-powered teamwork systems as holistic, integrated platforms to maximize the return on AI investments and create meaningful productivity gains at scale.
What to watch next
Tech leaders should evaluate their AI strategies by extending beyond development environments and integrating AI capabilities across all tools involved in software projects. Monitoring adoption and productivity metrics in asynchronous communication, aligned documentation, and decision tracking will be key indicators of success.
Future trends will likely include more AI features that summarize and contextualize knowledge across platforms, automate follow-up actions, and reduce interruptions, enabling engineering teams to focus more on creative problem-solving rather than administrative overhead. Organizations that adopt this systemic AI approach early may gain a competitive advantage in team productivity and product delivery speed.