The AI Productivity Gap in Project Management
AI project management tools are software systems that combine task tracking, communication, and automation with artificial intelligence to convert everyday work data into structured, coordinated workflows that promise faster, more predictable project delivery. Yet those promises are largely unfulfilled. Asana calls this the AI productivity gap: three quarters of knowledge workers now use AI on the job, but only five percent of companies report meaningful productivity gains. That mismatch shows that adoption alone does not equal impact. Many organisations run isolated pilots or bolt AI onto existing tools without changing how work is planned, assigned, and measured. The result is a layer of clever assistants sitting on top of the same fragmented processes, creating impressive demos but not fewer meetings, clearer ownership, or shorter project timelines.
Why Task-Level Time Savings Don’t Shrink Workloads
Individual workers describe compressing hours of effort into minutes with AI, from document drafting to meeting summaries and report generation. A business intelligence engineer at Amazon said AI cut document writing from over an hour to 15–20 minutes, while a Google security engineer reported that summarising six months of meetings now takes five to ten minutes instead of up to two hours. Yet they also say workloads feel unchanged. Time saved on one task is quickly redirected to the next project, or into building new automation systems that require upfront investment. Without changes to project scope or headcount expectations, AI workflow automation accelerates the pace of tasks but not the volume of demands. The AI productivity gap enterprise leaders face is therefore structural: they are optimising micro-tasks while leaving the overall flow of work, priorities, and capacity untouched.
From Prompt-Based Helpers to Workflow-Driven AI
Early AI project management tools centred on prompts: users asked for email summaries, draft documents, or code suggestions, and AI responded. Asana’s leadership argues that this model cannot plug into real-world business processes because it treats work as isolated requests rather than connected workflows. Their answer is what they call Agentic Work Management, built on an enterprise Work Graph that stores goals, decisions, ownership, and dependencies. Instead of waiting for prompts, agentic AI workplace tools act inside those structures: they identify who should do what, by when, and why. This shift from prompt-based chat to workflow-driven AI requires teams to rethink integration. AI needs to live where work is planned and tracked, not only where content is written or messages are sent. Otherwise, time savings stay local to individual tasks and never translate into shorter projects or fewer handoffs.
Dash and AI Teammates: Asana’s Bid to Close the Gap
Asana’s new AI chief of staff, Dash, and upgraded AI Teammates try to close the gap between AI capabilities and project outcomes by sitting across tools and channels. Dash is tied to an individual, watching meetings, email, and Slack threads for follow-ups and turning them into trackable tasks in the Work Graph. Each morning it can brief users on blocked items and pending decisions, then route work to the right AI Teammate with the human still in control. Teammates now come with reusable skills and integrations with tools such as Gmail, Outlook, Slack, HubSpot, Figma, and Canva, plus pre-built versions for industries like retail and manufacturing. Asana’s research claims that companies scaling AI are 43 percent more likely to report revenue growth than those stuck in experiments, but whether these new agentic layers deliver consistent time savings remains an open question for enterprises.
