Building SRE culture at a utility: the 18-month transformation.

The technical case for SRE in grid operations is strong. We made it across six articles. The economic case is compelling, with measurable returns at every utility scale. Neither matters if the organization can't adopt it. Utility culture presents real barriers to SRE adoption. They're also solvable barriers, because utilities already operate under frameworks that map directly to SRE principles. The transformation takes roughly 18 months. Here's how to execute it.

§ 01The cultural barriers

Utilities are among the most conservative organizations in the economy. That conservatism exists for good reason: mistakes kill people and collapse infrastructure. The same conservatism that prevents reckless decisions also prevents necessary adaptation. Understanding the specific tensions is the first step toward resolving them.

The risk-aversion resolution deserves the most attention. SRE doesn't reduce safety margins; it increases them through continuous validation rather than periodic testing. A utility that runs chaos experiments in a digital twin, as described in Part 04, takes on zero physical risk while discovering failure modes that periodic planning studies miss entirely. The framing matters. SRE isn't "move fast and break things." It's "understand exactly how things break so you can prevent it."

Siloed departments are the structural barrier. Distribution engineering, transmission operations, IT, cybersecurity, and vegetation management all affect reliability, but they typically report through different chains with different metrics. Shared SLOs force collaboration. When a SAIDI target requires coordinated action from operations, vegetation management, and field crews, the silo walls become visible obstacles rather than invisible defaults.

The seniority-to-meritocracy shift is manageable with dual career ladders. A senior engineer who masters SRE tooling and chaos-engineering methodology should be able to advance without moving into management. Google, Netflix, and every major tech company solved this decades ago. Utilities can adopt the same model.

§ 02Existing frameworks that map to SRE

Utilities don't need to invent SRE from scratch. Several frameworks already in use across the energy sector and adjacent industries map directly to SRE principles. The conceptual distance is shorter than most utility leaders assume.

Aviation Safety Management Systems

The aviation industry operates under Safety Management Systems (SMS) mandated by the FAA (14 CFR Part 5) and ICAO (Annex 19). The four SMS pillars — Policy, Risk Management, Assurance, and Promotion — have direct SRE parallels. Utilities frequently reference aviation as an analogous safety-critical industry. The mapping is direct.

Aviation's Safety Promotion pillar is particularly relevant. It establishes a culture where reporting incidents and near-misses is encouraged rather than punished. That's identical to the blameless postmortem culture SRE requires. If a utility already runs root cause analysis in the spirit of aviation SMS, the transition to blameless postmortems is a vocabulary change, not a cultural revolution.

High Reliability Organization principles

Many utilities already identify as High Reliability Organizations. The five HRO principles, defined by Karl Weick and Kathleen Sutcliffe (Managing the Unexpected, 2001; 3rd ed. 2015), map cleanly to SRE practices. The Midwest Reliability Organization and several NERC regional entities explicitly frame their operations around HRO principles.

1. Preoccupation with failure. The chaos engineering mindset. Rather than assuming systems work, HROs actively look for ways they might fail.
2. Reluctance to simplify. SRE demands deep, honest analysis of complex system interactions. Outage postmortems that stop at "a tree hit the line" fail this principle.
3. Sensitivity to operations. Maps directly to real-time observability. HROs maintain constant awareness of operational state.
4. Commitment to resilience. SRE's incident response capability — automated playbooks, practiced runbooks, chaos-validated recovery procedures — is the engineering implementation of organizational resilience.
5. Deference to expertise. SRE's meritocratic structure ensures the person with the most relevant knowledge drives the decision, regardless of title. During an incident, the engineer who understands the failing system leads the response.

If your utility already operates under HRO principles, you have the cultural foundation. SRE provides the engineering practices to make those principles measurable and repeatable.

NERC CIP standards

NERC Critical Infrastructure Protection standards are mandatory for bulk electric system operators. They also map to SRE practices with minimal translation.

CIP-009 is the most compelling entry point. Utilities already have recovery plans. SRE asks one additional question: have you tested them under realistic failure conditions? Chaos engineering in digital twins validates recovery plans without risking physical infrastructure. That positions SRE as a compliance enhancement, not a compliance risk.

§ 03Applying the Kotter model

John Kotter's 8-step change management model (Leading Change, 1996; updated 2012) provides a proven structure for organizational transformation. Applied to utility SRE adoption:

1. Create urgency. Aging infrastructure, escalating cyber threats, rising customer expectations, worsening outage statistics from climate-driven extreme weather. The data makes the case.
2. Build a guiding coalition. Operations, engineering, IT, and regulatory affairs all need a seat. Leaving out any one of them creates a veto point later.
3. Form a strategic vision. "A digital utility delivering 99.999% reliability through automation, continuous validation, and data-driven operations."
4. Enlist a volunteer army. Identify early adopters, designate SRE champions in each department, fund targeted training programs.
5. Remove barriers. Fast-track procurement for monitoring tools. Modify HR policies to create technical career ladders. Authorize digital twin experimentation without executive approval for each test.
6. Generate short-term wins. Automate a manual reporting process. Implement basic SCADA monitoring dashboards. Show the first SLO dashboard to leadership. Wins in the first 90 days build momentum.
7. Sustain acceleration. Expand SRE practices to additional systems. Implement chaos engineering in digital twins. Begin error budget tracking against regulatory metrics.
8. Institute change. Update job descriptions. Establish the SRE career path. Incorporate SRE metrics into performance reviews and rate-case filings.

§ 04The 18-month roadmap

Most utilities can execute this transformation in roughly 18 months. Faster is possible with strong executive sponsorship; slower is common when regulatory coordination is heavy. The phases below are sequential in the sense that each depends on the one before, but they overlap significantly in practice.

Months 0-3: Foundation

Build the guiding coalition. Name an SRE lead with direct reporting to the COO or VP of Operations. Conduct an SLI inventory — what are you already measuring, and how well? Select one pilot area (a single control center, one feeder cluster, or one asset class) to build initial capability. Start chaos engineering in a digital twin only. Don't touch production yet.

Months 3-9: Pilot and prove

Deploy the first set of SLOs on the pilot area. Establish burn-rate tracking and the error-budget policy. Run your first blameless postmortem on a real incident. Automate compliance reporting for IEEE 1366 indices. Build the first SLI dashboard and put it in front of executives weekly. Generate and publish the first short-term wins.

Months 9-15: Expand

Expand SLOs to 3-5 additional operational areas. Begin hardware-in-the-loop chaos testing. Create the dual career ladder and open the first technical-track promotions. Integrate SRE metrics into rate-case filings and NERC reporting. Start training across the broader engineering organization.

Months 15-18: Institutionalize

Move SLO dashboards into daily operations rhythm. Begin microgrid-scale chaos experiments with regulator awareness. Publish your first annual reliability report that includes SRE metrics alongside IEEE 1366 indices. Update performance review templates. Codify the error budget policy in the regulatory compliance manual.

By month 18 you have: measurable SAIDI and SAIFI improvements attributable to SRE practice, a trained cohort of SRE practitioners, a rate-case narrative that regulators reward, and an operational rhythm that compounds the gains year over year.

§ 05The people question

Utilities often worry they can't hire the talent. The reality is more nuanced. SRE isn't primarily about hiring software engineers. It's about teaching reliability engineers, system operators, and planning staff to work with SRE tools and disciplines. Most of the people you need are already on staff. They need training, tools, and permission to operate differently.

The hires you do need are narrower than the stereotype suggests: one or two data engineers who can build and maintain the SLI pipeline, one SRE practitioner with software background to lead tooling, and a reliability program manager who can coordinate across silos. Everything else is training your existing people to use SLOs, run postmortems, and design chaos experiments.

Training pathways exist. Google publishes the SRE books for free. Cloud providers offer SRE certifications. Utilities can partner with universities and technical colleges to build certificate programs tailored to the sector. This isn't a talent crisis. It's a training investment.

§ 06What success looks like

Eighteen months in, a utility that committed to this transformation has concrete, auditable results.

• SAIDI reduction of 20-40% on automated areas, tracked continuously against an explicit SLO.
• An error budget policy that governs operational decisions and has been cited in at least one rate-case filing.
• A blameless postmortem practice that has converted at least 10 incident reviews into documented systemic improvements.
• Chaos engineering in digital twins validating recovery plans that previously sat untested in binders.
• A trained cohort of 10-30 SRE practitioners distributed across operations, engineering, and compliance.
• A compliance automation stack that reclaimed half an FTE or more from manual reporting.

None of those outcomes is speculative. Utilities that have partially adopted these practices (EPB, ComEd, SMUD, several DOE SGIG participants) already show them. The full-stack adoption is still rare, which is exactly why the first utilities to commit will shape the next decade of rate cases and regulatory expectations.

§ 07The series, closed

Seven articles. One argument. The grid the industry was designed to run doesn't exist anymore. The grid we actually operate now is a networked, bidirectional, cyber-physical system that demands different reliability math, different operational rhythm, and different organizational behavior than the one the N-1/N-2 generation built.

The techniques to run that grid well already exist inside utility operations. FLISR, reclosers, protection testing, storm drills, black start. What's missing is the connective tissue: shared targets, burn-rate tracking, continuous measurement, error-budget discipline, blameless postmortems, chaos validation. SRE supplies that tissue. The economics work at every scale. The regulatory fit is direct. The cultural barriers are real but solvable.

The utility that moves first gets the first-mover advantage in rate cases, regulatory credibility, and operational reliability. The one that waits writes the postmortem.

If you want to talk about what this would look like at your utility, I'm at adam@sgridworks.com. The first call is always a diagnostic. Thirty minutes, and if I'm not the right person for what you're trying to move, I'll tell you who I'd call instead.

— Adam · adam@sgridworks.com · Apr 15, 2026