Brief IA
›

Asana: CEO Delegates Agenda and Oversight to AI Agents

AI Tools·Tom Levy·

Asana: CEO Delegates Agenda and Oversight to AI Agents

Asana: CEO Delegates Agenda and Oversight to AI Agents
Key Takeaways
1Dan Rogers, CEO of Asana, uses several AI agents daily to manage emails, calendar, clients, sales, and competitive monitoring
2The AI agents have safeguards: restricted access to sensitive data and limited actions without human validation
3Some agents communicate with each other without human intervention, and their effectiveness depends on precise instructions
Why it matters : Asana illustrates the advanced integration of AI agents in business management, with concrete applications and security limitations.

Asana's CEO, Dan Rogers, relies daily on a constellation of in-house AI agents. These bots handle emails, calendars, sales, clients, competition, and engineering capacity, with access limits and human validations for sensitive data. The agents can also communicate with each other and improve when the instructions are precise.

Safeguards and Agents Communicating Without Intervention

For sensitive information, Asana restricts access for the business analysis agent and frames actions without human approval. Dan Rogers emphasizes that security is paramount. On the monitoring side, the competitive agent produces daily briefings, which the CEO typically reviews every three or four days. This agent also interacts with an agent used by the product director: when a competitor launches a feature, they compare it to Asana's existing or planned offerings and flag anything that could lead to a change in the roadmap. According to Rogers, these exchanges occur independently, without him or the product director needing to initiate the conversation.

Prioritizing His Day with Three Digital Teammates

Each morning, Dan Rogers calls on three agents to structure his day. One analyzes his Gmail inbox to identify messages requiring a quick response, prioritizing those related to his goals and his team's projects. A second ensures that his calendar aligns with his priorities and alerts him if it does not. A third agent sifts through meeting transcripts to identify new commitments and update existing tasks.

Spotting Customer Signals and Adjusting the Sales Pitch

One agent scours Asana's customer database to surface points of attention such as delayed subscription renewals, and also collects complaints and requests posted online. Rogers consults it about once a week and explains that this information guides the choice of customers to contact to address issues before they escalate. The agent is customizable, down to the format of reports and sources, and submits changes to its creator that could benefit other users. On the sales side, an agent answers employee questions about products based on internal documentation and training materials. Rogers monitors its activity daily to identify areas of misunderstanding; if questions concentrate on a new pricing model, it signals to the company that it needs to explain it better.

Le brief IA que lisent les pros

Le brief IA que les pros lisent chaque soir

Les 7 actus IA du jour, décryptées en 5 min. Gratuit.

Inclus dès l'inscription : notre sélection des meilleurs guides & comparatifs IA.

Choisissez votre rythme

Gratuit · Pas de spam · Désabonnement en 1 clic

Tracking Performance Without Waiting for Financial Teams

A business analysis agent tracks sales results and the progress of new signings, aggregating internal systems to generate reports and dashboards on multiple key indicators. Dan Rogers notes that previously he had to request this information from the CFO or the business intelligence team, but he can now obtain it directly. Given the sensitivity of the data, Asana restricts the agent's access and the actions it can take without human validation.

Allocating Engineering Resources at the Right Place and Time

A capacity agent helps Asana decide where to deploy its engineering resources. When bandwidth remains available between two launches, it suggests undertaking smaller tasks. Rogers also uses it to track when customer-requested features might be developed. He indicates that this brings him peace of mind, appreciates seeing the progress of projects, and notes that the agent also satisfies his curiosity.

Effective Agents… When Instructions Are Precise

Rogers reports that the bots sometimes require detailed instructions to be truly useful. He cites a competitive intelligence agent that initially frustrated him by surfacing irrelevant information, before becoming effective after more precise framing. Work-oriented agents rely on institutional knowledge, a focus that Asana formalized in 2024 by opening collaboration with AI agents. Dan Rogers, 49, took over the company the following year and now entrusts these tools with part of his messaging, calendar, client relationships, and engineering priorities. Asana, founded in 2008 by Dustin Moskovitz and Justin Rosenstein to reduce work coordination time, offers a platform connecting team activities.

Brief IA — L'actualité IA en français

L'essentiel de l'actualité de l'intelligence artificielle, décrypté et expliqué chaque jour.