Your engineer works for you

You hire an engineer to answer a question, and too often what comes back is a study that has already assumed the answer. The board sees a professional stamp, a clean report, and a number, and the meeting turns into a vote instead of a conversation. Nobody asks why the fix costs that much, or whether a smaller version was ever on the table, because nobody in the room feels qualified to ask.

Rung 4 of the manager track.

That gap is not a technical failing. You do not need an engineering degree to manage an engineer well. You need to know what to ask for before the study starts, what a cost number is actually built from, and when it is worth paying for a second set of eyes.

The engineer works for you. Acting like it starts before the contract is signed.

Scope the work before you ask for a price

A vague scope produces a vague answer. A vague answer produces whatever project the engineer thinks is right. If your risk-ranked asset list already flags a well or a stretch of main as the priority, say so in the request.

Ask the engineer to study that asset, answer one specific question, and price one specific decision. Do not ask them to "look at the system" and tell you what it needs.

Name the decision the study has to support. Are you choosing between rehabilitating and replacing a well? Sizing a new storage tank?

Deciding whether a failing pump station needs a full rebuild or a targeted repair? A study written around one decision comes back with an answer you can act on. A study written around "assess our infrastructure" comes back differently: a list, a big number, and no way to tell which part of it actually matters.

Also name the accuracy level you are paying for, and put it in writing. A preliminary estimate is accurate to somewhere between 30 and 50 percent either direction. A parametric estimate, built from cost-per-unit formulas, tightens that to around 20 percent.

A detailed estimate is the kind used for final budgeting and funding applications. It needs real design work and contractor quotes, and lands within 5 to 15 percent. If the board is voting on a number, know which of those three numbers it is looking at.

What a cost estimate is actually made of

A construction estimate is not one number, it is three, stacked. Hard costs (materials and labor) typically run around 65 percent of a project. Soft costs (engineering and design, permitting, legal and administrative work, project management, testing and inspection) run around 25 percent.

Contingency covers the rest, usually 10 to 30 percent, higher early in planning and lower once the design is locked down. When in doubt, the rule is to plan the contingency high rather than low.

Knowing that split lets you ask a sharper question than "why does it cost so much." You can ask what portion is soft cost, and whether that lines up with the 25 percent norm. You can ask what contingency percentage the engineer used. Then ask whether the project is early enough in planning to justify a number at the high end of that range.

It also helps to carry a few reference numbers of your own, so an estimate does not arrive as a black box. Water main runs on the order of $65 per linear foot, more if the trench hits rock. Service line connections run $2,500 to $3,500 each.

A new well runs $200,000 to $500,000 depending on size. Storage runs $3 to $5 per gallon of capacity. A fire hydrant runs around $12,500.

None of these numbers tell you what your project should cost. Every system is different. But they tell you whether a number you are handed sits in a reasonable range or needs an explanation.

The questions that turn you from bystander to client

A client asks questions before signing off, not after the invoice arrives. A short list worth asking on any study that reaches your board:

• Did you consider repairing or rehabilitating this asset, or did the study default straight to replacement?
• What accuracy stage is this number, and what would it take to tighten it?
• What drove the cost up the most, and is that driver something we can change?
• Was a smaller or phased version of this project ever priced, or only the full build?

On pipeline work specifically, be alert to a common oversell. Trenchless methods (boring, lining, pipe bursting) get pitched as saving 30 to 40 percent over digging up the street. Real project data puts the actual saving closer to 10 percent. The genuine advantage of trenchless work is that the road stays open.

You also avoid tangling with other buried utilities underneath it. It is not that trenchless is dramatically cheaper. If a proposal leans on a big cost saving from going trenchless, ask to see the numbers behind that claim.

Ask what is actually driving the bill on any pipe project too, because it is rarely the pipe itself. Valves and fittings, service reconnections along the run, and surface restoration afterward typically cost more than people expect. Traffic disruption during construction is often the largest cost the public feels, even though it never shows up as its own line item. A study that only prices the pipe and skips these is not a complete price.

When a second opinion is worth the money

You do not need a second opinion on every project. Get one on projects where the dollar amount is large relative to your budget. Get one where the recommendation is a full replacement rather than a repair. Get one when your system has never faced this type of decision before.

A second, independent read on the same asset does one of two things. It confirms the first number, which is worth knowing. Or it surfaces an option nobody proposed, which is worth more.

Picking the engineer matters as much as getting the second opinion. The guidance from utility funding programs is blunt on this point. Hire the most qualified engineer for the type of project, not the cheapest bid and not simply the engineer you already know. A firm that has built ten wells in your kind of geology brings pattern recognition a generalist cannot.

Budget for the front-end work too. Funding applications for major projects often require a preliminary engineering report and an environmental review before anyone even considers the application. Both cost real money before you know if you are funded. Treat that spend as part of scoping the project correctly, not as a sunk cost to resent later.

Buying the diagnosis, not just the fix

The habit that separates a managed engineering relationship from a rubber stamp is paying for the diagnosis before you pay for the fix. One utility in Portland, Oregon, facing a failing pump station, was offered a full six million dollar upgrade. They had already tracked down which part of the station was actually failing. That let them replace just the failing controls for one million dollars instead of rebuilding the whole station.

That is the difference between paying for the offered work and paying for the right work. The offered work is whatever the engineer scoped by default. That is usually the safest, most complete version from a liability standpoint.

The right work is whatever your own condition data says actually needs fixing. You only get to ask for the second one if you already know your own system going in. That means scoping the work narrowly enough that a targeted answer was even possible.

None of this makes you the engineer. It makes you the client the engineer has to answer to, instead of the audience the engineer presents to. Next on the ladder is what happens after the money is approved and the project moves from paper to construction.

See after you win the money. Or return to the manager track for the full run of guides.