Case Study 13-1 — The $340,000 Question: Twenty Minutes on Bid Day

All companies, people, and projects in this case study are illustrative composites.

Setup

The project. The Ridgeline Regional Library and Learning Center — $31.4 million, 78,000 square feet, two stories plus a partial basement, for the Rivermont County Library District. Design-bid-build, lump sum, hard bid, public owner, prevailing wage, 100% payment and performance bonds. Bids due at the district's administrative offices at 2:00 p.m. A bid received at 2:01 is not a late bid. It is not a bid.

The wrinkle. At the sixty-percent design review, the county co-located a public health screening suite into the building program — three exam rooms, a small procedure room, and a records office. It is about 2,100 square feet of a 78,000 square foot building, and it carries medical gas: oxygen, medical air, and vacuum, with a manifold room, alarm panels, and third-party verification of the piping system.

Nobody expects medical gas in a library. That is the entire point.

The people. Tomás Reyes, Kestrel's chief estimator, is bid captain. Two estimators on quote intake. Dani Okonkwo, field engineer, on loan from Northgate for the day as scribe — writing down every quote with the time, the company, the number, and the stated exclusions, and doing nothing else. Nadia Haddad, Vice President of Operations, arrives at 1:00 for the final review, because at Kestrel nobody but Nadia moves the number.

The mechanical package. Kestrel's in-house detailed estimate for Package 23A is $5,120,000. Three bidders were solicited and all three confirmed they were bidding.

What happens

11:40 a.m. — the board

Kestrel's estimate sits under them at $5,120,000. Two real numbers, both above the in-house estimate by 4% and 6%. That is a normal, healthy picture — the market is a little hot on mechanical and Tomás already suspects his labor rate is a month stale.

11:52 a.m. — the third number

Ashfield Mechanical calls in at $4,978,000.

That is $340,000 under Cardinal and $142,000 under Kestrel's own estimate. On a $31.4 million bid in a market where the last three public jobs were decided by less than 1.5%, $340,000 is not a data point. It is the job.

Tomás does not celebrate. He says: "What did they miss?"

Ray Alvarez, on the practice: I want you to sit with the difference between "did they miss something" and "what did they miss." The first is a yes-or-no question and the honest answer is usually "I don't know," which lets you carry the number and feel fine about it. The second assumes the gap and makes you go find it. Tomás has been doing this for nineteen years and I have never once heard him ask the first version.

11:53–12:05 — the scope check

Ashfield's proposal is a single page. Dani reads the exclusions aloud, slowly, in order:

Excludes medical gas systems. Excludes test and balance by independent agency. Excludes duct cleaning and final filter change. Excludes cutting and patching. Excludes painting. Excludes fire alarm interface wiring. Excludes concrete pads and curbs. Excludes seismic bracing calculations by others.

Tomás stops her at the first one and pulls the bid package. Here is the comparison that decides the day:

Three of the eight exclusions are real holes. The other five — concrete pads, painting, cutting and patching, fire alarm interface, seismic calculations — are excluded by all three bidders and are carried elsewhere in Kestrel's estimate. They do not move the comparison.

Now do the arithmetic Tomás did in his head at 11:58:

$4,978,000 + $383,000 = $5,361,000

Against Cardinal at $5,318,000, Ashfield is $43,000 high. The $340,000 advantage never existed.

12:08 — the phone call

Tomás dials Ashfield himself and asks exactly three questions:

1. Is medical gas — piping, manifold, alarm panels, and third-party verification — in or out of your number?
2. Is test and balance by an independent agency in or out?
3. Are duct cleaning and the final filter change in or out?

He does not say what Cardinal bid. He does not say Ashfield is low. He does not say "you're going to need to be under five-three." He asks what is in the number.

⚖️ The line, in practice. Tomás's three questions change nothing about anybody's price by revealing anything about anybody's price. He is asking a bidder to describe their own scope. The moment he says "Cardinal is at $5,318,000, can you beat it?" he has crossed from clarification into bid shopping — and it would not merely be unethical, it would be a bad trade. A number pulled down by that pressure comes back out of the building.

12:26 — the revised number

Ashfield calls back:

Ashfield's adds come to $348,000 against Kestrel's carried $383,000 — they are $35,000 cheaper on the gaps, mostly because their medical gas partner is closer and hungrier. Their revised number is $5,326,000, against Cardinal's $5,318,000.

Cardinal is low by $8,000 — one-fifth of one percent.

1:00 p.m. — the final review

Nadia Haddad walks in and Tomás briefs her in ninety seconds. The mechanical carry is Cardinal at $5,318,000, $198,000 above Kestrel's in-house estimate.

Nadia: "Why is our own number $198,000 light?"

Tomás: "Two reasons and I can name both. Our sheet metal labor rate is from the last contract cycle — that's about $120,000. And I carried medical gas at $268,000; both real bidders are pricing it in the $240,000s, which means my number was fat there and light everywhere else. Net, my HVAC estimate was about four percent optimistic."

Nadia: "Do you want to carry Ashfield's revised number instead? It's a live proposal."

Tomás: "It's eight thousand dollars high and it was rewritten in thirty-four minutes on the phone. Cardinal's number has been sitting complete on my desk since 11:31, and Sofia Marchetti's people have run three of our jobs. Cardinal."

Kestrel bids $31,780,000.

The result

Kestrel loses by $412,000 — 1.3%.

Trask Builders carried Ashfield's original $4,978,000.

Fourteen months later, Ridgeline's medical gas scope is in mediation. Ashfield's position is that their proposal said "excludes medical gas systems" in plain English and Trask carried it anyway. Trask's position is that the bid package assigned medical gas to Package 23A and Ashfield acknowledged the bid package. Both are partly right, which is the worst kind of dispute. The screening suite is not finished, the county has taken partial occupancy of the rest of the building, and Trask has a superintendent and a project engineer still on site for a scope worth $268,000.

Analysis

What actually happened, mechanically

Nothing here was dishonest. Ashfield wrote an accurate proposal for the scope they priced. The failure was at the seam between two documents:

• MasterFormat is a specification structure. Medical gas is Division 22. Ashfield read the divisions.
• The bid package is a market structure. Kestrel put medical gas into Package 23A because it wanted one subcontractor accountable for the entire screening suite. Kestrel wrote the packages.

When those two structures disagree, only one document reconciles them: the scope sheet. Kestrel had one, and it is the reason Tomás found the gap in six minutes. Ashfield did not price against it carefully enough — and Kestrel's package narrative did not shout loudly enough.

Assign the fault honestly. It is not 100% Ashfield's. A bid package that reassigns a scope item across division lines must say so in a place a bidder cannot miss: in the scope sheet, in bold, in the package narrative, and in the pre-bid meeting minutes. Kestrel now does all four. That is what a lessons-learned process is for.

Why Kestrel lost by $412,000 and was right to

This is the part that is hard to teach and harder to live.

Kestrel had $383,000 of identified gap available on bid day. Carrying Ashfield's raw number would have put Kestrel at roughly $31,397,000 — within $29,000 of Trask's winning bid, close enough that a $50,000 adjustment anywhere else would have won the job.

So: Kestrel had the job available and declined it.

What Kestrel declined was not $383,000 of savings. It was $383,000 of unfunded liability with a fourteen-month tail, and the current state of the Ridgeline mediation is what that liability looks like from inside. Trask won a $31.4 million job and is now spending management time, legal fees, and extended general conditions on a $268,000 scope dispute, on a building the county has already partly occupied — which means Trask's leverage is gone and Ashfield's is not.

There is no version of this where Trask comes out ahead of where Kestrel is, having lost the bid and moved on.

The three habits that produced the result

1. "What did they miss," not "did they miss something." The first question assumes the gap and sends someone to find it. It takes six minutes when the scope sheet exists.
2. A scribe with one job. Dani's log — time, company, number, exclusions, verbatim — is what made the 11:53 scope check possible. Without contemporaneous exclusions written down as stated, the check becomes a memory exercise at 12:40 with three phones ringing.
3. Ask about scope, never about price. Tomás's three questions got Ashfield to correct their own number by $348,000 without a single competitor's figure leaving the room. That is what legitimate post-bid clarification looks like, and it is not a watered-down version of shopping. It is a different activity.

The counterfactual worth sitting with

Suppose Ashfield had come back and said "no, medical gas is genuinely in our number, we misdescribed it." Then Ashfield is low at $4,978,000, Kestrel bids about $31,397,000, and probably wins.

Tomás would have carried it — with a written confirmation, a scope sheet signed by Ashfield's principal, and an accepted-qualifications exhibit in the subcontract listing every exclusion Kestrel agreed to. The check is not designed to reject low numbers. It is designed to find out whether a low number is low or incomplete. Sometimes it is genuinely low, and then you take it and you win, and the check is what lets you sleep.

Discussion Questions

1. Kestrel's own detailed estimate for the mechanical package was $5,120,000 — 4% below both real bidders. Tomás named two causes on the spot. What should Kestrel do with that finding, and what is the risk of treating a 4% miss as "close enough"?

2. Ashfield's adds for the three gaps came to $348,000, while Kestrel carried $383,000. Ashfield was $35,000 cheaper on the same scope. Does that change your view of Ashfield as a bidder? What would you want to know before deciding?

3. Kestrel now flags cross-division package assignments in four places. Draft the sentence you would put in the scope sheet for the Ridgeline mechanical package so that no bidder could reasonably read medical gas as somebody else's scope.

4. Trask Builders won the job and is now in mediation over $268,000. Build the argument that Trask made a rational decision anyway — then rebut it. What information would settle the question?

5. Nadia asked "why is our own number $198,000 light?" before she asked anything about which bidder to carry. Why is that the more important question, and what does it tell you about how Kestrel uses its in-house estimate?

Your Turn

You are the bid captain on a $24 million hard-bid community aquatic center, bids due at 2:00 p.m. At 12:15 the low electrical number arrives at $3,210,000, which is $290,000 under the next bidder and $180,000 under your in-house estimate. Their proposal excludes: temporary power and lighting, fire alarm, the pool bonding grid, lightning protection, equipment pads, trenching and backfill for the site duct bank, and the utility company's connection charge.

Produce, in twenty minutes of your own time:

1. A table separating those seven exclusions into (a) real gaps against your bid package and (b) exclusions carried elsewhere or shared by all bidders. Assign a plausible value to each real gap.
2. The leveled comparison, and your recommendation on which number to carry.
3. The exact three or four questions you would ask on the phone — written out word for word, in a form that could be read aloud to the bidder without revealing anything about any competitor's price.
4. One sentence you would add to the electrical scope sheet on your next project so that this particular gap cannot recur.