A structured process for a complex problem
The process we designed had to do several things at once: surface the full range of AI opportunities without being captured by hype, evaluate them rigorously against business reality, and build genuine alignment among stakeholders who came to the table with different priorities and different levels of AI fluency. We began with comprehensive future scanning — mapping the key drivers and signals affecting the energy sector and its customers: legislative changes, technological shifts, social trends, and economic pressures. This wasn't background reading. It was the analytical foundation for everything that followed, ensuring the opportunity assessment was grounded in where the world is going, not just where it is now. From there, we ran systematic opportunity mapping across customer-facing and internal operations, using an impact-affect matrix to evaluate how each potential AI application would influence the organisation. Every opportunity was scored against both business value and organisational impact — creating a prioritisation framework that was transparent, comparable, and defensible to leadership.
The process was structured specifically to balance two things that often pull apart in large organisations: analytical rigour and stakeholder buy-in. Both matter. A direction no one owns doesn't get implemented; a direction without evidence doesn't survive contact with reality.
Direction, alignment, and the foundation for what's next
The core of the work was a series of strategic workshops with key stakeholders — leadership, domain experts, and data teams working together rather than in parallel. The workshops were designed to move from scanning to prioritisation to decision, with each stage building on the last. The focus for the AI direction centred on digital channels and customer experience, alongside opportunities in operational efficiency. By the end of the process, the organisation had a clear view of which use cases to pursue first, why, and in what sequence — with the analysis to support those choices and the stakeholder alignment to make them stick. The deliverables included an opportunity assessment and prioritisation framework the organisation could continue using beyond this engagement, and a strategic AI development roadmap aligned with the company's existing business priorities. The aim was not to hand over a report, but to build the internal capability to keep making good decisions as the landscape evolves.
The outcome was alignment — something harder to produce than a list of use cases, and more valuable. When leadership and domain teams agree on direction, AI initiatives move from exploration to execution.
Next projects.
(2016-25©)
