--- title: "Data Center Operations Leadership | Black Powder Search" description: "Role by role, what a critical facilities leader is accountable for during an incident, which parts of a Special Operations background map onto it, and where a…" url: https://blackpowdersearch.com/insights/data-center-operations-leadership --- [Insights](https://blackpowdersearch.com/insights) [Data centers](https://blackpowdersearch.com/industries/data-centers) # Data center operations leadership: what breaks at two in the morning Who should I hire to run data center operations, and does a military background actually help? By [Colin Brune](https://blackpowdersearch.com/team/colin-brune), Special Advisor September 19, 2026 Everyone in this sector has read that veterans are disciplined. It is true and it is not useful, because it does not tell you which veteran, for which seat, or what specifically they will do differently on a bad night. So here is the specific version, organized the way the problem actually presents: by what breaks, and who is holding it. ## The roles, and what each one owns when something goes wrong **Critical facilities manager.** Owns the physical plant — power, cooling, fire suppression — and owns the decision about whether to fail over. On a bad night this person is making a judgment with incomplete telemetry, under time pressure, where both action and inaction carry consequences. The failure mode is not panic. It is a plausible, fast, wrong call made because nobody established what was actually true before acting. **Field service or site operations lead.** Owns the people and the handover. Runs shifts, vendors, and the contractors who are on site the night something goes wrong. The failure mode is that a standard erodes when nobody senior is watching — and on a customer’s floor, nobody senior is watching most of the time. **Commissioning manager.** Owns the gap between what was designed and what was built. Runs the test sequence, and the job is substantially about refusing to sign off under schedule pressure from people more senior than them. **Electrical engineering manager.** Owns the design and the load, and the accumulation of small deviations that each make sense alone. These are four different jobs and they reward different things. The argument below applies unevenly across them, and any firm telling you otherwise is selling. ## What actually transfers ### Incident command as a rehearsed skill The thing that separates a good response from a fast one is a structure: establish what is actually true, decide with what you have, keep everyone working the same problem, and hold one person accountable for the decision. That is a rehearsed skill, not a personality trait. People who have it rehearsed it somewhere — and a Special Operations career rehearses it constantly, under conditions where the consequences were worse than a customer SLA. A Combat Controller or a Special Tactics operator has spent a career being the single point of technical authority inside somebody else’s team during exactly this kind of event. What you get, concretely: somebody who says “stop, what do we actually know” before anybody touches a breaker. That sentence is worth more at 2 a.m. than any certification on the resume. ### Handover discipline Uptime is lost at shift turnover far more often than it is lost to a component failure. What changed, what is degraded, what the next shift must watch — a formal, written, non-optional handover is the control, and it is boring, and it decays the moment nobody enforces it. Military operations are built on it. A leader arriving from that background treats handover as a real artifact rather than a courtesy, and will not accept “nothing much happened” as a turnover. This is the single most reliable transfer in the sector and it is almost never what anybody pitches. ### After action rigor, including after a good night A structured review held after an event covering what was supposed to happen, what actually happened, and why it differed — run regardless of outcome, with rank deliberately set aside inside the room. Two parts of that matter here. **Regardless of outcome** is the half most organizations skip: the near-miss that did not become an incident is the cheapest lesson available and it is invisible unless somebody looks for it. **Rank set aside** is what makes a technician willing to say the procedure is wrong. A leader with this habit installs blameless post-mortems without being asked. That is a cultural import worth more than the individual hire. ### Working on ground you do not control Colocation and customer sites mean operating under somebody else’s rules, alongside their people, with authority you do not have. An Army Special Forces career is built on precisely this — the core mission is working through partner forces rather than over them. This is the part that most often breaks a hire from a single-operator background, who is used to their own rules on their own floor. ## Where a career facilities hire is the better answer An honest version of this argument has to include the cases where the answer is no. **When the job is primarily design.** An electrical engineering manager owning a switchgear topology needs depth in the discipline, and depth in a discipline takes years of doing that discipline. Incident temperament does not substitute. **When the certification is the gate.** Some customers and some insurers require named certifications. That is a fact about your contract, not a hiring philosophy. **When you need somebody who has seen your exact failure.** There is no substitute for a person who has watched this specific class of equipment fail in this specific way. If your problem is narrow and recurring, hire the narrow expert. **When the role has no real authority.** This is the most common way these hires go wrong. Put somebody from this background in a seat with responsibility and no decision rights and you will lose them inside a year — not because they are difficult, but because a career spent being trusted to decide does not adapt well to being consulted. The honest shape is this: **the further a role sits toward operations, incident response and leading people, the better this population fits. The further it sits toward design depth and certification, the more you should hire for the discipline.** ## The hiring pattern we see work Pair them. A critical facilities leader from this background, with a deeply technical second who owns the discipline, is a combination that covers both failure modes. The leader will not pretend to know the equipment, and will build the structure that lets the specialist be heard at 2 a.m. The [roles we have filled in this sector](https://blackpowdersearch.com/industries/data-centers) cluster in exactly that shape — field service and site operations leadership rather than design authority. ## What to ask in the interview Beyond the [general twelve](https://blackpowdersearch.com/insights/how-to-interview-a-special-operations-veteran), three that are specific to this sector: - **“Walk me through the last time you were the one who had to call it with incomplete information.”** You are listening for the structure, not the outcome. - **“What does your shift turnover look like, in writing?”** If there is no artifact, there is no handover. - **“Tell me about a time a procedure was wrong and somebody junior told you.”** This tests whether the after action habit is real or claimed. The sector is short of people and the shortage is concentrated exactly where operations meets leadership. If you are filling one of those seats, [tell us the role](https://blackpowdersearch.com/for-employers#start) — the useful answer usually starts with which of the four jobs above it actually is. Next step ## Tell us the role. We will tell you who is already ready for it. An onboarding call with our COO and the lead recruiter, and you will know within the week whether the person you need is already in our pool. [Tell us the role](https://blackpowdersearch.com/for-employers#start) [See how we present a candidate](https://blackpowdersearch.com/talent) Keep reading - ### [Retained, contingent or outsourced: how to choose, and when to use none of them](https://blackpowdersearch.com/insights/choosing-an-engagement-model) Should I use retained or contingent search, and how do I tell which one my role needs? - ### [Hiring people who used to be the customer](https://blackpowdersearch.com/insights/hiring-people-who-used-to-be-the-customer) Should a defense technology company hire the people who used to be its end users? - ### [Twelve questions for interviewing a Special Operations veteran](https://blackpowdersearch.com/insights/how-to-interview-a-special-operations-veteran) What should I actually ask a Special Operations veteran in an interview, and how do I know whether the answer was good? Reference - [What triple vetted means](https://blackpowdersearch.com/triple-vetted) - [Reading a SOF résumé](https://blackpowdersearch.com/glossary) - [Data centers](https://blackpowdersearch.com/industries/data-centers)