Change is hard everywhere, but in highly regulated industries, it is unforgiving in a particular way. Processes are formal, expectations run high, and mistakes rarely stay internal. When a new technology stalls out inside a utility or a public agency, the reason is almost never the technology itself.
At LET'S GO! 2025, HData's annual customer conference, Jennifer Berrier, Executive Director at the Pennsylvania Public Utility Commission, joined Jason Stevenson, Senior Account Executive at HData, for a session on change management: what it actually takes to lead hundreds of users across multiple disciplines through a technology shift, inside an institution that operates under constant public scrutiny.
Berrier drew a hard line between the change management and project management: “Project management is where you're trying to achieve a specific project,” she said. “Change management is focused on people. It's focused on habits.”
To make the distinction concrete, she asked the audience to clasp their hands together, then reverse the position, interlacing their fingers the opposite way.
“That feels really awkward,” Berrier said. “So that's change. Change can be awkward.”
The job of change management, she added, is walking people from comfortable to awkward and back to comfortable again, without losing anyone along the way.
Before rolling out new technology, Berrier said her team focused first on earning buy-in rather than issuing a mandate. As part of that process, her team worked closely with Stevenson to test whether the HData tools would actually hold up in practice. “We came on-site and met with literally every single team,” Stevenson recalled. “If they said they wouldn't use it, we wouldn't have moved forward.”
That early feedback loop gave staff a voice before expectations were set, turning adoption into a shared decision rather than one imposed from above. “That very initial feedback of ‘is this even worth looking at?’ is a really important step,” Stevenson said.
One principle guided Berrier's approach at every step: Never assume people will accept a change before they understand it. “If you're being asked to do something differently, you want to know why,” she said. “You want to see the benefit.”
She compared the alternative to being told “because I said so,” an answer that rarely earns anyone's buy-in. Leaders, she said, need to explain the purpose behind a change, show how it makes daily work easier, and make the value concrete rather than abstract, then repeat that explanation as many times, and in as many ways, as it takes to land.
Change also cannot run on a fixed schedule, Berrier said. "You need to give people space to learn, space to train," she said. "At the end of the day, as a leader, you want to make sure your team has the tools they need to do the job effectively."
That includes confronting fear directly, especially the fear that a new tool is a threat rather than a help. "Assuring them that they're not going to be replaced or put out of a job as a result of technology is really big," Berrier said, pointing to the steady stream of unsettling AI headlines employees are absorbing on their own time.
Change does not move upward through an organization. It has to start at the top. “Active sponsorship by leadership is really critical,” Berrier said. “It has to come from the top. You have to champion it.”
At the Pennsylvania PUC, Berrier stepped in personally as the change's most visible advocate: setting expectations, modeling engagement, and making clear the shift mattered. Without that kind of visible leadership, she said, even strong tools struggle to gain traction.
Training at the commission involved staff going through two full days of hands-on sessions, tailored separately to attorneys, engineers, accountants, judges, and support staff, followed by a survey to learn what had resonated.
The more lasting move came after that: a power-user group embedded across the commission's bureaus. "These individuals are like our embedded trainers," Berrier said. "They love the product, and they can promote it within their bureau." Such an approach let knowledge travel peer to peer rather than only top-down, spreading adoption organically through the organization.
Adoption did not come at the expense of discipline. As a government agency, the commission operates under a statewide AI policy that explicitly prohibits AI from replacing human judgment.
"AI is not to replace employees," Berrier said. "At the end of the day, we want to make sure our highly trained staff are calling the shots."
In practice, that meant limiting use to approved tools, defining which uses were allowed and which were off-limits, and reinforcing at every turn that AI augments decision-making rather than replacing it. Those boundaries, Berrier said, made adoption safer, not slower.
Formal measurement at the commission evolved over time, eventually tracking training completion, usage statistics, and the types of tasks staff were using the tool for. But Berrier also pointed to something harder to put in a report.
"I like what I call the buzz," she said. "Hearing people talk about what they used it for today. That's an important metric." When people bring up a tool unprompted, she said, that is when you know change is actually taking hold.
The biggest misconception, Berrier warned, is believing change management has a finish line: "Change management does not end," she said. "It's a process that continues."
Technology evolves, teams turn over, and expectations keep rising. Guiding people through that is ongoing work. For leaders driving innovation, that may be the most important lesson of all.
Across the session, one point kept surfacing: Technology does not drive change on its own; people do. Approaching AI adoption with that in mind, Berrier and Stevenson's conversation suggests, is what turns a rollout into something durable.
Berrier and Stevenson's full full session and other videos at LET'S GO! 2025 are available for download. For more on AI built specifically for the regulatory work of utilities, regulators, advocates, and other energy professionals, visit hdata.com.
As the AI-native operating system for energy regulation, HData serves the largest customer ecosystem in regulated energy, helping utilities, regulators, advocates, advisory firms, corporates, and energy technology companies navigate regulatory complexity. Through centralized data, domain AI, and purpose-built applications, HData accelerates the research, analysis, and workflows critical to how the future of energy is decided.