Data can drive innovation and support growth in the digital age. An enterprise data strategy involves identifying ways to leverage data for new business opportunities. The key ingredient of an innovative data strategy is experimentation, and its not possible to experiment without having
- An environmental infrastructure for experimentation
- A strong data engineering foundation and support
- A culture that supports experimentation and innovation.
Environmental infrastructure for experimentation
The data strategy should support continuous experimentation of various use cases by creating an environment where data scientists can do extensive experimentation. This experiments should happen on real data, and need to be aligned with business goals. The experiments should constantly either prove or disprove various business hypothesis, and thus result in building use cases for improving products and services, reducing operating costs, reducing business risks, monitoring existing business models and creating new business models.
The environment where these experiments happen is popularly known as the data labs environment.
Strong data engineering foundation and support
Experimentation with data can only happen if the data scientists have the support of data engineers. Data scientists can experiment with data only if that data is available from reliable sources. Its highly inefficient to start hunting for data once we have identified a use case. Instead we need to have a data engineering foundation that will make all enterprise data available to data scientists, with sufficient quality and reliability, such that experiments on the data yield useful results.
Culture of Innovation
Although it may sound trite, fostering a culture of innovation is the biggest challenge. It not only requires investments and planning, but also requires a lot of patience, before the investments pay-off. It involves building competencies that are necessary for to do experiments, and requires the support and inclination from various business units. Business experts need to feed data and information to data scientists, so that they can build hypothesis and validate them with data. Data scientists on the other hand, need to have skills like stakeholder management and storytelling abilities, apart from technical and business skills.