The Low Bid That Wasn’t

Buyout. The job is won, the number is locked, and the project manager sits down to write the mechanical subcontract with the sub who came in $80,000 under the next bid. He opens the sub’s proposal to build the scope exhibit and reads the exclusions for the first time as a list instead of a line: housekeeping pads excluded. Controls wiring beyond the unit-mounted panel excluded. TAB re-balance after initial balance excluded. Everything else, per plans and specs.

He opens the electrical sub’s proposal. Controls wiring excluded — that is a mechanical item. Equipment pads by others. He opens the plumbing sub’s proposal. Pads by others.

Every air handler, pump, and water heater on the job sits on a concrete pad, and the pad is in nobody’s contract. The wiring between every field device and the building automation panel is in nobody’s contract. The mechanical bid was $80,000 lower than the next one, and now he can see exactly why — and the $80,000 is already gone, because it was the reason the GC’s number won.

Why it happens

Sub bids get compared to each other, and the comparison hides the one thing that matters.

On bid day, three mechanical proposals land within an hour of the deadline. The fastest way to compare them is to line up the totals, and the lowest total becomes the baseline every other bid gets measured against. That is the structural failure. When the low bid is row zero, an exclusion only shows up if the other bidders included it. If all three subs excluded the housekeeping pads — and they all do, because it is not a mechanical item and pricing it dulls the number — the comparison never shows a pad at all. The scope gap is shared, so it is invisible. The estimator is comparing three numbers that all left out the same thing, and the number that left out the most looks best.

The proposal language finishes the job. Per plans and specs reads like coverage and covers nothing, because it itemizes no scope. Silence on an item — no mention in inclusions, no mention in exclusions — gets read as included, because under deadline that is the only reading that lets the estimator move on. And the sub is not cheating. Excluding the pad is rational: it is another trade’s concrete, carrying it makes the bid uncompetitive, and if it surfaces later it surfaces as a change order to the GC, not as a loss to the sub. Three subs made the same rational move. The orphan they created has no owner because the specification put it between trades and the bid comparison never had a row for it.

What it costs

At bid, the pads and the controls wiring are two lines. Normalized into the comparison — added to every bid that excluded them at a historical unit cost or a supplier quote — they would have shown the $80,000 gap was closer to nothing, and the low mechanical bid was the middle one. The GC carries the orphan items as its own line in the number, or qualifies them on the bid form, and the number that wins is a number that covers the building.

At buyout, the orphans belong to the GC at whatever they cost now. The concrete sub has demobilized and is not coming back for pads at bid-day unit pricing. The controls wiring gets negotiated with a mechanical sub who knows the GC has no other option and prices accordingly. The TAB re-balance surfaces at commissioning as a change order nobody can push back on, because every sub can show the line that excluded it. And the $80,000 that made the GC’s bid the winner was never real — it was scope, not savings, and it moved from the sub’s number to the GC’s fee the moment the subcontract was signed.

Catching it early

The gap is visible at bid, but only if the baseline is the specification instead of the cheapest proposal.

Build row zero from the spec before the bids arrive: Division 01 requirements that bind the trade, the trade’s own technical section, and the coordination requirements in adjacent divisions that create scope the trade section never mentions — the pad in Division 03, the controls wiring sitting between Division 23 and Division 26. Translate the requirements into line items the matrix can score. Then score every line for every sub as one of four things: included, excluded, qualified, or silent. Hold the last two apart from the first. Qualified means conditional. Silent means unknown. Neither one means carried, and per plans and specs is silent for every line it does not name.

Cross-reference the exclusion lists across every trade and look for the items every sub excluded. Those are the orphans, and they are findable in the proposals before the bid closes. Normalize every exclusion into dollars — historical cost, a supplier quote, a unit cost database — so the matrix produces a true cost per sub instead of a bid price per sub. On bid day, triage: level the highest-value, highest-exclusion-risk trades first, make the verification calls on the exclusions that move the number, and for the gaps that cannot be resolved before the deadline, carry the orphan as a GC line item or write the bid form qualification that names the trade and the item. Then hand the whole matrix to buyout, so the subcontract exhibit incorporates the leveled scope instead of the bid-day proposal.

That is what Gauge does. It builds the matrix with the specification as row zero, scores every sub bid against it, finds the items every sub excluded and no one carried, and hands the estimator a true cost per sub before the number locks. Every estimator levels against the low number. Gauge levels to the spec — which is the only place the low bid that wasn’t can be seen for what it is.