Skip to main content

πŸ“ Gauge β€” Sub Bid Leveling

Four bids on one trade. One is $250K under. Hero, or missing scope?
Gauge reads your specification, builds the required-scope list for the trade, and scores every sub bid against it β€” included, excluded, qualified, or silent β€” with each gap priced and the spec section cited. The award stays yours.
Run it on a package you already closed. You know the answer β€” see whether Gauge finds it.

The Problem

Package eight of eleven closes in forty minutes. The mechanical number is $250K under the next guy, and there are two possibilities: he is sharp, or he skipped the commissioning. The proposal will not tell you which. That is what the proposal is for.

A hundred and fifty PDFs and one clock. The qualifier that rewrites the scope is on page 38, in the paragraph nobody reads β€” and at 9pm, after two hours of moving numbers into a spreadsheet, the fresh eyes that would catch it do not exist.

Then it surfaces on its own schedule. Estimators on the submitting side describe making the call themselves: “I would get call from GC’s to tell me my number is out of whack and may have missed something. Sure enough, I did.” The miss was catchable. The comparison just could not see it.

Your own peers wrote the diagnosis on the review sites: subcontractor pricing is “not apple to apple 99% of the time.” Every tool in the category promises apples to apples. None of them deliver it, because all of them are comparing the bids to each other.

The Stakes

If nothing changes, the gap survives the award β€” and after award it stops being a phone call. The sub will not pull an exclusion without an offsetting increase. That is not bad faith; it is the number they priced. At buyout the same gap costs a price increase. In the field it costs a change order at change-order rates, and the month-four conversation where a project manager asks why the trade that came in lowest is now the most expensive line on the job.

One part does not reverse: the award itself. In one documented case, a GC missed an excluded-conduit line, awarded on the low number, and paid twice β€” once for the change order, and once in the better-known contractor it could have had at a lower true price. The comparison that would have shown it did not exist on bid day.

Who it helps

Estimators closing trade packages on a deadline, who know the low number is probably hiding something and do not have time to read page 38 of every PDF to find out what.
Preconstruction managers on guaranteed-maximum work, who inherit an award that went to the bid that skipped the most scope and have to claw it back before Notice to Proceed β€” off the fee if they cannot.
Chief estimators who built the leveling template the whole office trusts, and know the exclusions tab is thinner than it looks.
Operations project managers handed a bought-out job, reconstructing what each subcontractor is actually carrying before a trench, a wall assembly, or a roof penetration turns into a crew standing on contested scope.
Subcontractors who price the whole specification and keep losing to the bidder who skipped it β€” when the general contractor levels against the spec, that discipline stops being a competitive liability.

How It Works

Nothing here touches a live bid. Run the first one on a closed package, alone, before anyone else in the office sees it.

  1. 01
    Pick one trade and hand over what is already on your desk: the spec sections, the addenda, and the bids.
  2. 02
    Gauge builds the required-scope list and scores every bid in four states β€” included, excluded, qualified, silent.
  3. 03
    You get leveled totals with every adjustment traceable: each gap priced, each basis stated, each section cited, each open question flagged instead of guessed.
  4. 04
    You make the award. The record goes with the job.
β€œSection 23 08 00 requires commissioning. None of the four proposals mention it, and none of them excluded it β€” so none of them priced it, and all four believe somebody else is carrying it. Your total goes up today with a section number next to it, or it goes up in month four with a change order number next to it.”
Gauge

What it does

Gauge is a bid leveling specialist, and the difference is where the baseline comes from. Every category on this market builds the comparison from the bids, so an item no bidder mentions never enters it. Gauge builds the required scope from your specification first β€” Division 01, the trade section, the adjacent divisions, the addenda β€” and then the bids become verdicts against it.

01

Silence is a scored state.

The $250K mystery is usually a silence: not included, not excluded, just never mentioned. Gauge scores it, cites the section that requires it, and prices it β€” so the bid that skipped the most scope stops looking like the bid that priced it sharpest.
02

It reads the prose.

The qualifier on page 38 is not a number, which is why no column comparison surfaces it. Gauge reads the exclusion and qualification language and flags the ones that rewrite scope, with page and section.
03

Every add-back carries its basis.

The section, the assumption, the source of the number β€” so the leveled total survives challenge in the award meeting and can be defended by someone who did not build it.
04

The record travels.

Every verdict, citation, and priced gap goes into buyout as a document a stranger can read β€” so the negotiation opens with a number and a justification instead of an archaeology project.

And two things it does not do. It does not replace your leveling template β€” it extends coverage into the exclusion tab everyone knows is thin, inside the structure your office already uses. And it does not award. Gauge front-loads the read and cites it; the judgment, the award, and the risk stay with the estimator.

The Promise

With the specification sections in hand, here is what Gauge will produce: every requirement the spec carries that no bidder priced, flagged with the section cited and a defensible dollar on it β€” and wherever the documents genuinely do not answer, a flag naming what would resolve it, instead of a confident guess.

What that looks like when it works: you walk into the award meeting and say this is the real low number, and here is why β€” and the why is on paper, in a record you can hand to the person above you without staking your credibility on a tool’s say-so. The reasoning is visible, which means it is checkable, which is the point: a promise about how often a tool is right does not tell you what happens on the occasion it is wrong. A citation does.

What Gauge will not promise: that a sub surrenders an exclusion after award β€” they will not, without a price. That change orders disappear β€” some are owner-driven and design-driven, and Gauge touches neither. Or that your read becomes unnecessary. It becomes cited.

Objections

β€œWe already have a leveling tool.”

You paid for this once already, and your peers wrote down what happened: the module tabulates numbers that are not comparable and calls it leveling. Tabulation displays what bids state. Leveling scores them against what the project requires β€” including the requirement no bid mentions. Ask your current tool to produce the silent row. That row is the product.

β€œCan it actually read the qualifications?”

The right way to answer that is to watch it miss or catch on your own material. Run a package where you know a qualifier rewrote the scope, and watch it get caught, priced, and stated with its basis. If the read is wrong, you will see where and why β€” the record restores your authorship of the judgment instead of asking you to trust a percentage.

β€œNot during bid season.”

Correct β€” nothing about this belongs on a live clock. The evaluation is retrospective: a closed package, a known answer, run by you alone. Bid season is when the results get used, not when the tool gets learned.

β€œI built our template. I know what it does. It stays.”

Load your scope sheets; the structure your office trusts remains the structure. What changes is coverage β€” the exclusion tab extends to the full required scope of the spec, on every package, whether or not anyone had time to build it that deep. Your standard gets stronger. It does not get replaced.
The bid that skipped the most scope looks exactly like the bid that priced it sharpest β€” right up until it is scored against the document.

Pick a package you already closed and run it.

Known answer, no live clock, nothing installed, your template untouched. If Gauge finds nothing you did not already know, you have spent one closed package confirming your read. If it finds the silence you missed, you found it before it had a change order number attached.

Every solution in this market compares bids to each other, which means a scope item nobody mentioned is invisible to all of them.