How Leaders Make AI Work for Real Teams: Lessons from Asana’s COO

Leaders aren’t just adopting AI tools — they’re rethinking how work gets organized in the first place. In her conversation with PMI CEO Pierre Le Manh on The Shift Code Podcast, Anne Raimondi, COO and Head of Business at Asana, explains how the same forces that fueled SaaS adoption — ease, transparency, and measurable value — are now shaping how organizations introduce AI. Executives want proof that tools improve outcomes, and Raimondi sees customers using AI to cut down on “work about work,” strengthen collaboration, and create more room for deeper thinking and creativity.

Goodbye “Work About Work”

Raimondi uses a simple phrase for the friction every team feels: “work about work.” It’s the coordination, clarification, and follow-up that slows progress without improving the outcome. She highlights intake requests that arrive incomplete, long back-and-forths to gather missing details, and manual steps that delay work unnecessarily.

AI is already reducing some of that drag. Customers are using automation to improve requests before they reach a teammate, apply established guidelines to streamline reviews, and shorten workflows that once took weeks into processes completed in days. The gains show up in quality as much as speed, because AI can draw on prior best submissions and existing patterns to make work clearer from the start. These improvements depend on something deeper than automation alone; they require context.

Why Context Is the AI Superpower

For Raimondi, context is what makes AI genuinely useful. Asana’s work graph connects people, tasks, projects, goals, and permissions, giving AI the structure it needs to understand how work fits together. Instead of producing isolated outputs, AI can strengthen requests, reveal blockers, and tie tasks to the goals they support.

Transparency is essential to making that context trustworthy. Teams need to understand the steps an AI system took so they can refine it when needed. And as organizations consolidate tools for security and data reasons, having a single source of structured context becomes even more important. Context shapes how AI performs , but people shape how AI is used. That’s where the human side of hybrid teams comes in.

What Human/AI Teams Actually Look Like

Raimondi emphasizes that organizations aren’t eliminating whole roles; they’re handing off specific steps of work that follow predictable patterns — such as intake triage, first-line support, initial content drafting, or translation. These shifts are most visible in early-career positions, where routine tasks make up a larger share of the workflow.

At the same time, AI opens up creative possibilities. People often hesitate to ask a colleague for dozens of variations of a headline, but they’ll freely iterate with an AI system. That iterative space helps teams explore more options before applying human judgment and brand expertise.

Employees also want clarity on how AI will be used: which tools are supported, how their roles might evolve, and what outcomes leadership expects. That foundational understanding is critical for hybrid teamwork to function well. And providing that clarity — along with governance and oversight — is where leaders play a decisive role.

Why Accountability Still Belongs to Humans

As AI becomes embedded in daily workflows, accountability still belongs to people. Employees are unsure who is responsible when an AI system makes a mistake — the person who deployed it, IT, the application vendor, or the model provider. Clear governance helps eliminate that uncertainty. Leaders need policies that specify which AI capabilities are approved, when humans must stay in the loop, and how outputs are reviewed before deployment.

Transparency is a core part of this system. Actions taken by an AI system should be auditable, so teams can see how a recommendation was generated and offer targeted feedback. Raimondi shares an example from Asana’s communications team, where a leader trained an AI teammate to help prepare briefing materials. Because the feedback was visible, the rest of the team could observe her reasoning and strengthen their own work in the process. These themes point toward a practical set of behaviors leaders can adopt.

A Practical Playbook for Leaders Introducing AI

Raimondi observations point to several practices that help organizations adopt AI effectively:

Set clear guardrails.

Define supported tools, appropriate use cases, and when humans must remain in the loop.

Connect work to goals.

Link tasks and projects to measurable objectives so both people and AI systems can prioritize effectively.

Make decisions transparent.

Ensure teams can trace how AI-recommended actions were generated and intervene when needed.

Redesign workflows where repetition is highest.

Start with predictable steps — intake, initial drafts, QA — where AI can reduce back-and-forth.

Measure the impact.

Track cycle-time reductions, quality improvements, and hours saved to understand where AI creates value.

These leadership practices position organizations to benefit from AI — and they hint at how business models may evolve alongside them.

How AI is Reshaping the SaaS Value Model

Raimondi sees AI reshaping the economics of SaaS. Subscription will remain, she explains, but more value will come from consumption — the compute required to power AI capabilities. That model ties cost more directly to measurable outcomes.

Looking ahead, she envisions organizations managing two essential resources — people and compute — with greater transparency and intention. The aim isn’t just faster execution, but work that helps people feel more connected to purpose and impact. For Raimondi, that’s the promise of an AI-enabled enterprise: clearer goals, better alignment, and tools that help teams achieve what matters most.

Tags: Artificial Intelligence | Future of Work | Leadership | Innovation

Related Insights

You May Also Like