A successful data science project is one that results in implementation, e.g., putting a model into production or using the result to support an organizational change. But implementation is only partially a technical problem. Successful implementation depends on people. This talk will cover multi-level stakeholder analysis, and techniques for engagement from project scoping to delivery of results.
A successful data science project is one that results in implementation, e.g., putting a model into production or using the result to support an organizational change. But implementation is only partially a technical problem. Successful project implementation depends on people.
This talk will cover how to set up your project for success from the very beginning, including conducting a multi-level stakeholder analysis, establishing expectations through a well-defined communication cadence, and directly involving key stakeholders in problem definition and scoping by applying design thinking and other collaborative techniques borrowed from management science and social science.
We will also cover several use case studies (identities withheld to protect the innocent!), drawn from real-world experience in client-facing analytics, where these techniques have been successfully applied. This talk is useful for practicing data scientists of all levels, whether in academia, on an internal analytics team, or client-facing.