How to Turn a Business Process Into an Agentic Workflow

font-style: normal; font-variant-ligatures: normal; font-variant-caps: normal; font-weight: 400; letter-spacing: normal; orphans: 2; text-align: start; text-indent: 0px; text-transform: none; widows: 2; word-spacing: 0px; -webkit-text-stroke-width: 0px; white-space: normal; background-text-decoration-thickness: initial; text-decoration-style: initial; text-decoration-color: initial; display: inline !important; float: none;">Kreijkes stressed the importance of understanding that clients often don't know what they need from AI. "The client almost never knows what they want because AI is new," she explained. To close this gap, it's crucial to ask probing questions and involve various stakeholders in the conversation. This approach helps uncover inefficiencies that the business owner might overlook. By engaging with different team members, you can identify repetitive tasks suitable for automation, ensuring that the AI solution addresses the most pressing issues.

Using Visualization Tools

Visualization is a powerful tool in designing AI workflows. Marralissa recommended using Figma to map out processes, which helps clients visualize the workflow and understand their role within it. This method prevents miscommunication and sets clear expectations. "When you write it down, your clients become creative," she noted, emphasizing that a visual representation can spark new ideas and ensure everyone is on the same page. This clarity is essential for avoiding scope creep and ensuring that the delivered solution meets the client's needs.

Key Components of AI Workflows

An effective AI workflow should include personalization, self-improvement, and performance analysis. Personalization ensures that the AI solution is customized to the client's specific needs, while self-improvement mechanisms, like feedback loops, help the system evolve over time. Performance analysis, using tools like N8N's evaluations, allows for ongoing assessment of the workflow's effectiveness. Marralissa highlighted the importance of these components, stating, "Your automation should support, and it should not complicate."

Engaging Stakeholders with the Crazy Eight Exercise

To encourage creativity and uncover innovative solutions, Marralissa introduced the Crazy Eight exercise. This involves gathering stakeholders to brainstorm potential AI applications, encouraging them to think outside the box. By setting a timer and allowing free-flowing ideas, participants can generate a wide range of possibilities. This exercise not only surfaces unique solutions but also ensures that the final workflow aligns with the needs of the entire organization, not just the business owner.