Atlassian: AI Boosts Individual Speed, But Not Organizational Efficiency
Dr. Molly Sands, head of Atlassian's Teamwork Lab, argues that companies are mismanaging AI adoption by focusing on individual productivity rather than team collaboration. Her research, which involves behavioral scientists and psychologists, aims to reshape how work is done within organizations by studying AI's impact on teamwork. A recent Atlassian report surveyed 12,000 global knowledge workers and interviewed 200 Fortune 1000 executives, revealing a significant gap: while 89% of executives noted individual employees are faster with AI, only 6% could identify specific ROI. Only about 14% of teams have successfully translated AI use into tangible value, suggesting pockets of high performance within otherwise stagnant organizations.
Sands identified three key characteristics of successful teams: context, workflows, and culture. These teams build a 'context graph' by documenting goals, decisions, and knowledge in shared digital spaces, allowing AI to access necessary organizational information. They also redesign entire workflows instead of just optimizing isolated tasks, preventing individuals from working at cross-purposes. Furthermore, these leading teams foster a culture of experimentation and learning, explicitly accepting that some initiatives may fail. To facilitate this, Atlassian suggests leaders implement AI working agreements, defining AI usage, acceptable tools, and common skillsets to ensure shared understanding and context.
Sands contends that AI is not introducing novel management challenges but rather amplifying pre-existing issues like hidden assumptions and differing mental models. The technology highlights the critical need for shared context and explicit work processes. By imposing deliberate constraints and encouraging experimentation, teams can accelerate learning and identify effective AI applications. The adoption of AI working agreements has shown promise in increasing AI usage, improving decision-making, and enhancing work quality.
AI's current impact on organizational productivity appears to be a classic case of optimizing local efficiencies without addressing systemic bottlenecks. While individual task acceleration is a visible outcome, the lack of corresponding organizational ROI suggests that core issues in collaboration, knowledge sharing, and workflow design are hindering broader gains. The 'context graph' concept points to a need for more robust digital infrastructure that captures and disseminates institutional knowledge, moving beyond individual cognitive reliance. Future organizational success will likely hinge on designing AI integration strategies that prioritize collective intelligence and adaptable workflows, rather than solely focusing on individual tool proficiency. This shift requires proactive leadership to redefine team structures and foster environments where experimentation and shared learning are systematically encouraged and supported.
AI-generated to prompt reflection — not editorial opinion, not advice, not a statement of fact. How this works.