Artificial intelligence offers tremendous potential to strengthen your business and that of your customers. However, it requires foresight, discipline and a plan to make it a reality. There is a gap between organizations that use their data effectively as a strategic asset and those who don’t. This is due to the rapid pace of change. According to a 2019 Wall Street Journal article, most corporate executives believe that poor data quality will hinder their AI efforts.
Although it may seem obvious to use data to make decisions, many organizations still have process and organizational silos that prevent them from effectively leveraging their data.
Experian also found that 83% of organizations view data as an integral part their business strategy, while 69% believe inaccurate data continues to hinder their efforts.
Building blocks to unlock data success
Inconsistency and data silos can lead to inefficient processes and hidden costs that can result in poor decision making and missed opportunities. Organizations looking to leverage artificial intelligence (AI/ML), as data is the fuel that runs the algorithms, can compound these inefficiencies. Too often, data scientists or engineers spend significant time and effort cleaning data before an AI/ML modeling can be used. This can lead to missed opportunities, increased costs, and longer timelines.
The AI Advisory Council at CompTIA has identified six key building blocks that will enable better data usage and unlock future, advanced capabilities like AI/ML throughout the enterprise.
Align corporate strategy and data: Although it sounds obvious, many companies fail to align their corporate strategy with technology and digital strategies. This helps you to identify the gaps and prioritize your priorities. AI/ML is undoubtedly the most pervasive technology. Corporate strategies today should be able to clearly understand how AI/ML can help drive value across the enterprise and where AI/ML may disrupt their business.
Prioritize use cases: Similar as the first point, strategic alignment should include a number of prioritized use cases that unlock business capability. Organizations with a strong data infrastructure should explore AI/ML use case to learn more about what is possible. As the organization matures, more valuable use cases and tangible results will become apparent. If an organization doesn’t have the foundation, it should focus on a series of foundational projects to unlock these desired capabilities. Executive alignment and funding should be made clear by the interdependencies between advanced capability uses cases and foundational use cases.
The right team should be assembled: Organizations should review their current team and consider a people strategy. This could include reusing/repurposing existing resources, training, upskilling/training, or hiring third-party skills. Some organizations may only require their third-party IT partner to help them model data and implement the software. As an organization grows, there will be a variety of roles that you need to consider in a data-driven culture, including data scientists, data engineers, and others. Based on the organization’s size and needs, new groups may be needed to support the data driven culture. An example of this is an AI center for excellence or a cross-functional automation team.
Define your architecture: Packaged and commercial software have advanced data capabilities. They are also starting to embed AI within their applications. Business intelligence platforms can also be used to provide a complete data management solution, depending on the needs of an organization. Cloud platforms are a key enabler for AI solutions. They offer more processing power and storage. It is important to have an architecture that can manage software delivery in order to ensure consistency and extensibility.
Define data owners and boundaries: Access to the right information at a time is the ultimate goal of efficient IT systems. A data strategy must be developed that supports the corporate strategy. There are many data artifacts that can help you define your data ownership, data usage, and other aspects of data management. Although large organizations may have a central data or analytics group that is responsible for data management, the most important task is to map the data ownership within each organization. Data steward is one of the new roles and titles that organizations can use to understand the governance and discipline needed to make a datable organization.
