What we self-perform

The number, and what it actually means

We self-perform more than 75% of the work on a typical project with our own workforce and our own equipment. That figure is the most load-bearing fact on this website, so it is worth being precise about what it means and what it does not.

It means that on a normal water or wastewater project, more than three quarters of the contract value is executed by people on our payroll, using plant we own or hold, under our own supervision. It is not a measure of how much of the contract we hold — any general contractor holds all of it. It is a measure of how much of it we do.

What it does not mean is that we do everything. Nobody does, and a contractor who claims otherwise is either misdescribing subcontractors as crews or doing specialist work badly. What we subcontract, and why, is set out below.

Why it matters more on a live plant than anywhere else

On a straightforward job, self-performance is mostly a cost argument: fewer layers of mark-up. On a plant that has to keep meeting its permit while it is rebuilt, it is a risk argument, and the risk argument is much stronger.

Here is the mechanism. A retrofit sequence is a chain of dependencies where the plant is exposed for a defined period, and the length of that exposure is set by the slowest link. When the excavation belongs to one firm, the concrete to a second, the pipe to a third and the site work to a fourth, every handoff between them has float built into it — not because anyone is lazy, but because each firm is protecting itself against the others being late. That float accumulates, and on a retrofit it accumulates directly into the number of days the plant has no margin.

When the same four scopes belong to one crew list, the sequence compresses to what the work actually needs. The excavator that opens the trench is the one that backfills it, the crew that formed the wall is the one that strips it, and there is no negotiation about who goes next.

The second mechanism is accountability, and it is the one owners notice. When something goes wrong across a handoff between two subcontractors, an owner gets two defensible accounts of whose delay it was and a change order. When it goes wrong between two of our own crews, there is nobody to blame and nothing to argue about. That is an uncomfortable position to be in and it is the correct one.

The third is the tie-in. At two in the morning, with a basin out of service, the people doing the work should be people whose training we know, whose previous tie-ins we have watched, and who report to the same superintendent. That is not a cost argument at all.

What our own crews do

Earthwork and excavation, including trench support and dewatering. Cast-in-place concrete, including water-retaining structures, formwork, reinforcing placement, finishing and curing. Pipeline installation in open cut — water main, gravity sewer, force main — with testing and disinfection. Manholes and structures, precast and cast-in-place. Process and yard piping. Selective demolition inside operating facilities. Bypass pumping installation and attendance. Site civil: grading, drainage, plant roads, erosion control and restoration. Mechanical equipment setting, alignment and grouting.

Those are the scopes where the schedule risk on a plant project lives, and they are the ones we do not hand to anybody.

What we subcontract, and why

Electrical and instrumentation, because a licensed electrical contractor and a controls integrator are specialist trades and should be. We hold the interface and the commissioning responsibility, but the work belongs to people who do only that.

Trenchless crossings — bore and jack, directional drill, pipe bursting — because these are equipment-and-operator specialisms where the firms who do them daily are better at them than a general civil crew, and the failure modes are unforgiving.

Protective coatings in immersion service, where surface preparation and holiday testing are certified trades. Paving and striping, which is a production business with its own plant. Structural steel fabrication. Surveying and materials testing, which should be independent of the contractor as a matter of principle rather than convenience.

Where we subcontract, we carry the coordination, the schedule and the responsibility to the owner. The subcontractor is not an explanation an owner should have to accept.

How to check the claim rather than take it

Self-performance is the easiest thing in this industry to claim and one of the easiest to verify, so verify it. Nothing below requires our permission.

Ask for certified payroll on a comparable completed project and count the trades on it. On a genuinely self-performed job the payroll carries operators, carpenters, cement finishers, labourers and pipe layers. On a brokered job it carries a superintendent and a project manager.

Ask for the equipment schedule — owned, not rented. Excavators, loaders, dozers, trench boxes, pumps, plate compactors and trucks are capital, and a firm that self-performs owns them because renting them by the month is how you lose money.

Ask who the superintendent on your job will be, and ask to talk to them rather than to the estimator. Then ask the last three owners whether that person was actually on their site.

Ask for the subcontractor list at bid time and compare its value against the contract value. That single division is the whole claim, and it is a number, not an adjective.

What it does not fix

Self-performance does not make a firm the right size for every project, and it is worth saying that plainly on a page that is otherwise an argument for it.

A crew list is a fixed capacity. It means a well-matched project gets our own people and a badly-matched one would get subcontractors, which is the thing this page says is worse. So the honest version of the claim includes a scale: if a project is large enough that we would have to broker most of it, we are the wrong contractor for it and we would rather say so at the pre-bid meeting.

It also does not substitute for the sequence. A self-performed retrofit with a badly planned bypass is still a badly planned retrofit. The two arguments are independent, and the sequence is the one that keeps the plant in compliance — see bypass, shutdown and tie-in.