Striking a Harmony in Prioritizing Client Demands and Data Scientist Insights
In a unique collaboration, a data science and technical product team joined forces with a client product team to develop machine learning and analytics dashboards for device health management. This cross-functional client team, consisting of product managers, business stakeholders, data/analytics experts, technical stakeholders, and end users, worked closely with the technical team to create effective, user-friendly dashboards that met both business goals and technical feasibility.
At the heart of this project was the client product manager, who guided feature prioritization and ensured the product aligned with the client's strategy and objectives. Business stakeholders provided strategic direction and funding approvals, while data/analytics experts validated data-driven insights and ensured their relevance for business decisions. Technical representatives handled integration with existing IT infrastructure, data sources, and security constraints, and end users provided essential feedback on usability and business value.
The key to the project's success was the co-manager model, which assigned equal authority and responsibility to both the client and technical team. This approach created a sense of ownership for both teams in the project's success and enabled a continuous feedback loop, aligning analytics outputs with actual client business needs and device health management goals.
Initially, there were conflicts between the two parties due to unmet expectations and goals. However, improved explainability and communication led to clients taking more responsibility for the project's direction. The client's product manager was empowered to take on more responsibility and make decisions based on their business case and desired outcome.
The dashboards were created for a specific subset of devices, narrowing the project scope. The UI was easy to navigate and provided detailed explanations in the client's terminology, which further empowered clients to aid in decision-making and ownership of the product they support.
The team followed the Four Principles of Explainable Artificial Intelligence in their work, ensuring clients understood the data science processes and could contribute to the roadmap's success. Weekly meetings were held to discuss the project status, next steps, and demos, with explanation accuracy being evaluated based on client feedback during these sessions.
In a previous role, the technical team worked with a client who needed help with machine learning and analytics for their dashboards. This experience laid the foundation for the current project, demonstrating the value of a collaborative approach in delivering practical, effective solutions.
In the end, the project's goal was to produce device health management dashboards that provided accompanying evidence or reasons for all outputs, allowing analysts to click down to the underlying data for explanation. This approach not only streamlined decision-making but also ensured that the insights generated were grounded in concrete data, fostering confidence in the recommendations made.
This collaboration extended beyond device health management, as the data science team also worked on home-and-garden projects, leveraging data-and-cloud-computing technology to create personalized lifestyle suggestions based on individual usage patterns and behavior.
Moreover, the client product manager's role expanded beyond device health management, overseeing the integration of technology into various domains of the client's lifestyle, ensuring a cohesive and user-friendly experience across home-and-garden, data-and-cloud-computing, and other ecosystems.
