Connecting Data to Enable Innovation
Problem
A $1B original equipment manufacturer’s R&D department developed their own data and reporting architecture to enable their department to increase and mature its data capabilities faster than the rest of the enterprise. This approach eased some short-term pain points and allowed them to achieve fast turnaround times in their process design. As they scaled, challenges became evident. Driven by a lack of governance and myopic view of data management limited R&D’s ability they were limited in their ability to automate reporting, they saw eroded trust in their data, and this resulted in heavy utilization of key engineering resources on data tabulation rather than the critical role they needed to play in innovation of the company’s products. The R&D team recognized their architecture was creating a capacity constraint for their engineers and wanted to enable scalable analysis through fitting in to the enterprise’s data architecture, identifying low value analysis to be automated, and enabling their databases to be the source of truth.
Approach
Excelerate worked closely with a cross-functional group consisting of members from Business Intelligence, Manufacturing Engineers, Product Engineers and Research Engineers. Through a series of deep dive sessions focused on integration and automation, a key conclusion was that by making some simple data collection updates and enacting some foundational data principles, a new architecture would allow them to achieve their goals. With a focus on achieving stakeholder alignment for the changes that would enable capacity for engineers, Excelerate co-created a new governance structure that could be maintained with existing resources and would increase data integrity and integration across the complete product lifecycle.
Results
The focus on aligning Business Intelligence (IT) resources with R&D’s needs allowed for the development of a new schema that will align with the enterprise architecture, integrate their disparate data sources for a holistic view of product development, and eliminate the need to manually check data quality for all analyses. The resulting roadmap provided them with a phased approach that would maximize the gains from these changes early (within 3 months), followed by additional work effort to further optimize key data structures.