655: AI ROI: How to get a profitable return on an AI-project investment — with Keith McCormick

Clear Goals

Setting clear, revenue-oriented goals is crucial for AI project success. Keith McCormick emphasizes that transparency in the process allows clients to understand and embrace AI models, ensuring they align with business objectives 1. He suggests that AI projects should not be open-ended but structured with specific revenue goals from the start 2. McCormick explains, "If you're going to make a prediction, this is going to sound terribly old school, but I think that, generally speaking, we should be trying to predict yes, no questions" 3.

   

Org Structures

Effective organizational structures are key to maximizing data science project success. Keith McCormick argues that data science teams should be led by someone with C-suite influence to secure necessary resources and drive value 4. He highlights the importance of distinguishing between roles like Chief Data Officer and Chief Analytics Officer, as they focus on different aspects of data management and analytics 4. "I really want the CAO to be laser focused, solely focused on identifying which projects have value," he asserts 5.

   

Project Balance

Balancing exploratory and revenue-generating projects is essential for data science teams. Jon Krohn shares his approach of blending projects to maintain productivity while allowing for creative exploration 6. Keith McCormick suggests viewing data science teams as internal consultancies, balancing billable and non-billable time to ensure profitability 7. He notes, "Predictive analytics teams should be self-funding, they should be profit centers, not cost centers" 7.