Dan Domkowski February 2, 2022
An Architecture for Accelerating and Scaling Data-Driven Outcomes
Data is worthless. Well, not always, but the data you’ve acquired carries little meaning without delivering results like creating a competitive advantage or new market opportunities, detecting fraudulent patterns, or improving the quality of human life. It’s the potential for a positive impact on your business or mission that gives data value; data must lead to an outcome.
The democratization of advanced analytics mechanisms like machine learning is providing us with more pathways to positive data-driven outcomes. Those that can effectively harness their data in new and exciting ways can be thought of as building “opportunity factories” on behalf of their organizations. Nevertheless, the abundance of data sources, the variety of data formats, the increasing degrees of data diversity, and the ever-surging volume of generated data make it difficult to define repeatable patterns for all industries and domains.
Research and surveys alike have shown us that data science teams devote significant time and resources to the collection and preparation of data before they even get to analysis and model training. In response, we’re seeing evidence that high-performing organizations are putting an emphasis on reusability by taking a platform approach to data science, data engineering, and application development.
These trends tell us that a mix of desired behaviors and the right platform architecture, when executed together, can create powerful outcomes from data across your organization. Platforms provide a technology model for teams to share and reuse common components and capabilities that lead to repeatable patterns for success. We wrote our original whitepaper on Generalizing the Architecture for Digital Platforms and how they create positive leverage out of a shared business and technical framework.