Running an Operation That Survives Turnaround Season

Industrial service work does not arrive evenly. It arrives in bursts — a turnaround date that moves twice and then lands, a capital project that starts three weeks early, a named storm that empties the schedule and refills it two weeks later with entirely different work. Between the bursts, headcount shrinks back to a core that knows everything. During them, it doubles with people who know almost nothing about how this particular company does things.

That pattern is the defining operating fact of the business, and it is rarely designed for. Instead the surge is absorbed by the few people who remember how last season went. The company survives on memory, and memory is the single most expensive operating system available, because it cannot be in two places, it takes holidays, and it eventually resigns.

What actually breaks first when a crew doubles?

Not capacity. Decisions.

Doubling a crew is a hiring and equipment problem, and most companies in this line of work are reasonably good at it — the callout list gets worked, the agency gets called, the trucks come out of the yard. What breaks is everything that used to be decided informally because there were only nine people and all of them were within earshot of the owner.

Who says yes to a customer asking for two more bodies on Thursday. Who prices it. Who confirms the new hire’s site access is current before he is dispatched. Who notices that the same equipment got promised to two jobs. In a small crew those questions never get asked out loud because the answer is always the same person. At double the size they get answered by whoever is standing there, and the errors do not show up until billing or, worse, at a gate.

What has to be written down before the surge, not during it?

The short list is the decisions that will otherwise be improvised under time pressure, and it is shorter than most owners expect.

Write down the callout order and who is actually on it, with certifications and expiry dates attached, so the list is a roster and not a group text. Write down what mobilizing looks like at each customer site: access requirements, orientation, badging, whatever their onboarding demands, and how long each takes — that lead time is the difference between a crew working Monday and a crew waiting in a parking lot Monday. Write down the agreed pricing and the open commitments per customer, where the field can see them. Write down who to call when something goes wrong at two in the morning, and who makes that call.

None of this is a software project. It is a handful of documents that make the surge executable by someone other than the person who has done it for nine years. If nothing in the operation is written down yet, that is the actual first project, and the case for it is in SOPs are how small businesses scale.

Who is authorized to commit a crew?

This is the single most common gap, and it is worth answering explicitly rather than culturally.

An authorization matrix takes an afternoon: who may accept new work, up to what value, on whose terms, and what has to come back to the owner regardless of value. It should cover accepting a verbal scope change on site, committing people beyond the current schedule, agreeing a rate that is not on the rate sheet, hiring, and renting equipment. Each line names a role, a limit, and what the person has to record when they use it.

The reason this matters more during a surge than at any other time is that the cost of asking goes up exactly when the cost of guessing does. A superintendent who cannot reach the owner at 5 a.m. will make a decision anyway. The matrix makes that decision a delegated one instead of an unauthorized one, and it is the difference between a business that can absorb a second concurrent job and one that cannot.

How do you absorb a customer’s standards without carrying four systems?

By maintaining one internal standard set to the strictest requirement you have accepted, then mapping it outward to each customer’s format.

Every prime contractor and plant owner brings its own safety program, reporting cadence, documentation structure, and portal. Accepted one at a time, each looks like a condition of the contract, and it is. The accumulation is the problem: after four customers, crews are trained four ways, the office maintains four sets of records that contain most of the same facts, and nobody chose any of it.

The consolidation is a recordkeeping exercise, not a judgment one, and the distinction matters. What a training record must contain, which qualifications a task requires, and whether work is safe to proceed are determined by qualified people and by the customer’s own requirements — CMA is not a safety, environmental, or compliance firm, and no automation makes those calls. What can be built is the layer underneath: each fact captured once, held with its expiry date, and rendered into whichever format a given customer demands. Winning a fifth contract then means adding an output, not issuing a fifth rulebook to people already on site.

What do subcontractors and pass-through costs do during a surge?

They become the place margin leaks, because they are the part of the operation that scales fastest and is documented least.

When the schedule exceeds the crew, the difference is bought: subcontract labor, rented equipment, freight, third-party services. Each arrives with its own paperwork, its own insurance certificate that may have lapsed, and its own idea of what a daily record looks like. The practical rule is that a subcontractor’s record has to arrive in the same shape as your own, because anything else gets reconciled by hand later, and later is when the surge is over and everyone has moved on.

Pass-through costs need the same discipline on the way out. Costs incurred on a job and never rebilled are not an accounting oversight; they are work performed at a loss. Where those costs disappear is covered in from field ticket to invoice.

What has to keep working when a storm scatters the schedule?

Four things, and all four are decided in advance or not at all.

Who has the authority to demobilize and on what trigger. Where the current roster and contact details live, reachable without the office network. How standby, demobilization, and remobilization time is recorded and which of it is billable under each contract. And what documentation the business will need afterward — condition photographs, hours, equipment locations — which is far easier to capture while it happens than to reconstruct for an insurer in November.

An operation that rebuilds all four from scratch every season spends the scarcest currency of a storm week, which is attention. Writing them down once turns a scramble into a procedure a second person can execute.

Is this an operations problem or an owner problem?

Usually both, and the second one is the harder admission.

Every symptom above resolves to the same structure: the owner is the index of the business. Schedule, pricing, customer relationships, crew capability, and the history of what was agreed with whom all live in one head, and that arrangement works until two jobs overlap. The tell is not chaos — it is that the business cannot grow past the owner’s calendar, which is the argument made in why you need a COO before you think you do.

A fractional COO is one answer: a senior operator inside the business part time, owning how the operation runs rather than any single job, at $6,000 to $12,000 per month with a three-month minimum. It is not always the right answer. If the gap is running one specific large job well, an experienced project manager is the better hire. If the gap is purely scheduling, a dispatcher or planner is cheaper and more direct. The comparison against a permanent hire is laid out in fractional COO versus full-time COO.

It is also worth knowing what the current stack costs before deciding the fix is another subscription; the stack cost calculator totals it. Where several rented tools each hold a slice of the same job, an operations platform you own can cost less over the life of the business than what it replaces.

CMA is based in Dallas and has no office or staff in Corpus Christi. Travel for the on-site portion of an engagement is scoped and priced up front, because the yard, the dock, and the gate are where an operating cadence is either real or theoretical. What that looks like in this market is on the Corpus Christi fractional COO page.

The useful time to do this work is between surges, which is the time it never feels urgent. If the last turnaround season ran on memory, book a 30-minute call before the next one starts.