A Roadmap to Data Readiness
OWNERSHIP
STRATEGY
DATA CULTURE
Data readiness must be treated as a utility-wide strategic priority, with clear executive accountability for data strategy, governance, and investment decisions. Without clear C-suite alignment and support, individual business units may pursue their own data-readiness strategies, leading to duplication of effort and inflated costs. “Data needs to be used across the organization for AI. If there is a framework for data readiness in one business unit that is different in other business units, there’s a lack of efficiency, and you miss the force multiplier effect,” Wakefield said. “Having a top-down strategy helps the entire organization.”
Data readiness is ultimately defined by the specific business outcomes utilities want AI to support. Clarity on what those outcomes are makes the steps towards data readiness more targeted and efficient. This includes defining how data supports measurable business outcomes; ensuring data is well-documented and traceable; and identifying any data gaps that need to be closed.
Data has always been important to utility operations and decision-making. But AI has dramatically raised the stakes, demanding that utilities evolve accordingly. “Part of this is recognizing that data is an asset, just like physical assets,” Wakefield said. “When data is truly an asset, you need to be able to leverage it as an asset by ensuring it is accurate, accessible, and governed. Building a data culture means having the processes and mindset to do that.” Building a data culture requires continuous education about data literacy and fostering a collective understanding about the connection between data and the business outcomes utilities want to achieve.
Though the applications of AI are potentially transformative, their implementation demands significant change. Building the tools, processes, knowledge, and commitment for that change requires early successes. Starting with small use cases that deliver quick and tangible value builds internal support. “We selected use cases with high business value and manageable data complexity to demonstrate early success and build momentum,” said John Brady, senior manager for IT at Constellation Energy. For example, one use case is to leverage AI to evaluate drone imagery of transmission and distribution infrastructure. Properly trained AI can trigger predictive maintenance that repairs or replaces equipment before it fails. But those wins only build momentum if utilities actively communicate them across the organization.
SCALE
USE CASES
Demonstrate Value with Targeted Use Cases
Build a Data Culture
Build a Strategy That Connects Data to Business Outcomes
Secure Executive Ownership
Data readiness is not a task that can be completed quickly. It takes time, flexibility, and continuous learning. As deployments occur, utilities should capture and share any lessons learned and develop best practices to guide future initiatives around data readiness and data governance. Over time, this builds on existing data governance and management practices, increasing maturity and confidence and making it more likely that AI consistently delivers the outcomes utilities seek.
Learn, Scale, and Institutionalize
PREVIOUS
NEXT
Achieving data readiness doesn’t just occur naturally. It requires direction, leadership, and a commitment to continuous improvement. EPRI and its collaborators in this research developed a five-step framework to secure the data readiness needed to drive AI benefits and scale. The five steps are: