More and more businesses are relying on their ability to convert data into actionable insights to help them achieve their digital goals. Unfortunately, most eventually have to hit the brakes on their digital ambitions due to a lack of maturity in their data foundation and the overall data culture.
The challenge is an expensive legacy data warehouse purpose-built for specific reporting needs. Only a few in the Business Intelligence IT team were able to access the data and deliver ever-evolving insight needs of the business. The process was slow while the existing governance process made it difficult to open up access to a wide variety of people.
A newly formed data science team took matters into their own hands and built a data lake in Google Cloud Platform. Applying lean engineering principles, the team quickly stood up the data lake and an analytics sandbox and successfully started bringing AI-enabled experiences to market. The challenge quickly became managing the DevOps process. Should the data science team also support the DevOps process?
Once the business saw rapid results, the questions became how does the rest of the organization tap into this data lake? Should this become the enterprise data lake? How would this work?
Speeding up the building of the modern data platform foundation
Productive Edge began working with the Business Intelligence IT team to craft an enterprise roadmap for the adoption of the data lake.
Multiple areas had to be thought through. Starting with redefining the mission of the BI team and the creation of an iterative roadmap aligned to lighthouse use cases, Productive Edge further assisted in redefining data governance policies, data integration, best practices for data curation and eventual access of the data from the data lake.
Creating a data governance framework that focused on providing trustworthy data and that opened up access was key. Productive Edge assisted in establishing data governance policies and identifying data governance tools specifically for master data, data catalog, and data virtualization with a goal of opening data up while meeting regulatory and compliance requirements.
Data integration and the process for staging raw data, data cleansing and curation within the data lake were put in place. A master data strategy and approach was identified and aligned with the overall roadmap.
Finally, an operating model was put in place to have IT take ownership of MLOps and DataOps so that teams could focus on the analytics task at hand.
Jumpstarting a data-driven culture
The journey towards a modern data platform has begun. Because of its iterative nature, the impact is rapid and calculated. Organizational culture is changing and the entire company is invigorated with the knowledge that data can be harnessed to drive differentiation.