We scan new podcasts and send you the top 5 insights daily.
Jason advises picking one outbound or CX agent, pushing it as far as possible on a subset of work, then adding another only when gaps appear. He warns against buying every tool in their stack, says best-of-breed still matters, and admits they broke their own add-one-remove-one rule.
Attempting to remember and maintain different domain-specific personal agents across a company creates friction. When a single agent becomes exceptionally good at core tasks, employees naturally gravitate toward extending that centralized agent rather than building out isolated personal bots.
Resist building complex, multi-agent systems from day one. Instead, start with a single agent and build its skills based on actual workflows. Add sub-agents only when a clear productivity need arises. This approach is more effective than scaling for what looks impressive.
Start with a 'Minimal Useful Agent' that performs a simple, bounded task like drafting replies for human approval or triaging inbound requests. This 'draft and approve' model reduces risk, builds customer trust, and allows you to earn autonomy over time.
AI agent platforms are typically priced by usage, not seats, making initial costs low. Instead of a top-down mandate for one tool, leaders should encourage teams to expense and experiment with several options. The best solution for the team will emerge organically through use.
Moving past chaotic "hackathons," effective AI implementation needs a designated leader who knows the team's processes inside and out. This person shepherds the strategy, ensuring agents are built on a solid foundation and integrated smoothly, preventing a proliferation of uncontrolled, low-quality bots.
To avoid chaos in AI exploration, assign roles. Designate one person as the "pilot" to actively drive new tools for a set period. Others act as "passengers"—they are engaged and informed but follow the pilot's lead. This focuses team energy and prevents conflicting efforts.
Onboard users (or yourself) to an AI agent like a new human teammate. Start with easy, high-frequency tasks (e.g., summarizing Slack threads). Progress to harder, multi-step tasks (e.g., scheduling a meeting based on replies). Only then, attempt to automate an entire workflow (e.g., running daily growth experiments).
Instead of using simple, context-unaware cron jobs to keep agents active, designate one agent as a manager. This "chief of staff" agent, possessing full context of your priorities, can intelligently ping and direct other specialized agents, creating a more conscious and coordinated team.
Avoid creating too many specialized agents initially. Instead, have your 'Chief of Staff' agent perform a new task once successfully. Only after you've validated the process and its value should you 'earn the right' to create a new, dedicated agent to own that function.
Instead of a broad AI overhaul, CMOs should identify their most acute pain point in the inbound funnel—like slow lead follow-up or poor event lead conversion. Deploying an AI agent to solve that specific, high-impact problem first builds momentum, proves value, and de-risks wider adoption.