Initial success with AI workflows led to hundreds of unmanageable, invisible automations. Sendoso pivoted to creating role-based AI "digital colleagues" with org charts and managers, making AI implementation transparent and manageable for the human team.
Marketers are overwhelmed acting as the manual link between 15,000+ tools. The agentic marketing model, proposed by Kana, makes AI the central nervous system, connecting data and systems to automate complex decisions and executions, freeing up human marketers for strategic work.
An AI engineer knows how to build agents but lacks domain knowledge, while a subject matter expert knows what a great outcome looks like but can't build. Pairing them in an "agentic pod" is the key to creating high-performing, specialized AI agents that deliver real business value.
Sendoso found that building AI agents was technically easier than managing the cultural shift required for team adoption. Getting people on board, managing expectations, and creating new workflows proved to be the most critical and difficult part of their AI journey.
Before deploying AI agents, Sendoso discovered they needed more than just clean data; they had to build a complete data dictionary and ontology. Agents can't interpret ambiguous field names or tribal knowledge, forcing a foundational data cleanup and definition process.
Sendoso's initial strategy of giving every employee AI tools resulted in chaos: duplicate agents, inconsistent quality, and permission issues. They reversed course, centralizing development into an "Agentic Pod" model to maintain control, quality, and strategic focus.
Instead of building a generic marketing agent, audit your team for process gaps, underwater functions (like field marketing), or roles you lack the budget to hire. Your first AI agent should be a targeted solution to one of these specific, high-pain problems.
Research reveals a conflict: 40% of leaders believe the Chief AI Officer should own agentic marketing, while CMOs feel it's their domain. Marketers must demand a seat at the table during AI evaluation to prevent technology decisions that ignore critical marketing expertise and workflows.
Sendoso found that creating a central AI team relieved employees of the pressure to self-learn AI development. This stopped "usage theater," where people felt compelled to constantly use AI to seem productive, and instead allowed them to collaborate as subject matter experts.
A key mistake is applying AI to automate a process designed for humans. Instead, teams should redesign the process from first principles, leveraging the unique capabilities of AI agents, such as 24/7 operation and instant data processing, to unlock new efficiencies.