Chapter 13 Quiz — Detailed Estimating

C. Architectural precast genuinely belongs to Division 03 under MasterFormat — Kestrel filed it correctly. The glazier's number is legitimate for their trade. The failure is that the package ("the exterior wall") spans Divisions 03, 05, 07, and 08, and only a written scope sheet tells each bidder which of those they are being asked to price. Divisions organize the specification; packages organize the market.

B. $2,900,000 ÷ 565 = $5,132.74/CD, and Kestrel and Meridian negotiated a round $5,150/CD into the supplementary conditions. A stipulated rate trades a little accuracy for the elimination of a fight on every delay claim. The $10,650/CD in option C is the total daily exposure — $5,150 extended general conditions plus $5,500 liquidated damages — which is a different number for a different purpose.

C. Roughly $2.28 million of $2.9 million — about 78% — is time-dependent: every staff line, trailer rent, temporary power, dumpsters, portable toilets, vehicles, security, and photography. The one-time items are trailer setup, fencing, temporary roads, final cleaning, mockups, and the as-built survey. This is the mechanism behind "a schedule slip is automatically a cost overrun." Nobody has to decide to spend that money; it spends itself.

C. Bright Line at $2,778,000. Bright Line needed only $30,000 of adjustments; Cordova needed $429,000 and a bond, finishing at $2,974,822; Meridian needed $284,000, Addendum 4, and a bond, finishing at $3,002,132. The apparent low bidder ended up $196,822 above the apparent high bidder. Bright Line's number was not high — it was complete.

C. A shared exclusion is invisible on the bid tab — adding the same number to all three does not move anybody's ranking. But it is absolutely a scope gap, and on Northgate it must be covered in Division 13 with the radiation-shielding contractor. Shared exclusions are the most dangerous kind precisely because the leveling process is blind to them. Circle them separately, on their own list.

C. Bond premium is a rate on the full contract value, including the fee. This creates a small circular reference — the fee affects the contract value, which affects the bond, which affects the cost, which affects the fee — that estimating software resolves automatically and hand-built spreadsheets often do not. Two practical consequences: a change order that adds $1,000,000 of work adds roughly $8,250 of bond premium you must remember to include, and a bond computed on the cost of work instead of the contract value is an error that survives every review because the rate is defensible.

B. Escalate what you have not bought, for the period until you buy it. At GMP execution Kestrel had firm quotes covering about 34% of the cost of work; the exposed 66% is $26,400,000, split into $18,400,000 of material exposure at 2.30% ($423,200) and $8,000,000 of labor exposure at 1.90% ($152,000). Escalating scope already locked by a firm price is double-counting — roughly a $190,000 error in the owner's disfavor, and a competent owner's estimator will find it.

C. $1,804,800 ÷ $47,500,000 = 3.80%. (As a margin on the pre-escalation price of $46,924,800 it is 3.846%.) A true 4.0% margin would have required a price of $45,120,000 ÷ 0.96 = $47,000,000 and a fee of $1,880,000 — $75,200 more. Markup is added to cost; margin is a share of price; margin is always smaller.

C. Depreciation is charged per hour of use, so it does not change ($18.25/hr either way). But interest ($9,375/yr), taxes, insurance, and storage ($5,000/yr) accrue on the calendar and must be recovered over whatever hours the machine actually works. At 1,000 hr/yr that is $14.38/hr; at 600 hr/yr it is $23.96/hr. Utilization, not purchase price, decides own versus rent — which is why fleet decisions are company decisions, not project decisions.

B. Bid shopping — the GC disclosed (in effect) a competitor's price level and invited a lower number for unchanged scope. Bid peddling would be the same transaction initiated by the sub ("just tell me what to beat"). The test that separates both from legitimate negotiation: on the legitimate side, the scope or the terms change and the price follows. Here, only the price moved.

C. The loss is $998,000 − $784,000 = $214,000. At a 3.5% fee, recovering $214,000 requires $214,000 ÷ 0.035 = $6,114,286 of additional work performed at full fee with no overruns. A forty-minute decision to shave $465,000 created a $6.1 million hole. That ratio — not moral disapproval — is the argument against buying work.

D. A, B, and C all have a name and a mechanism: a real number replaced a plug, a named risk was retired, a schedule change reduced a time-dependent cost. Each can be written down and defended six months later. D is buying the job — reducing the number without a corresponding change in scope, risk, or expectation. It is legal, it is common, and it is the most reliable way to lose money in this industry.

False. Write the row and put a dash in it. A dash means "I looked and it is zero." A missing row means nothing at all, and six weeks later nobody can tell a decided zero from a forgotten scope.

False. The cost of the work and general conditions are reimbursable at actual cost and are typically auditable. The fee is not — it is a negotiated fixed amount or percentage with nothing behind it to inspect. That distinction is exactly why the boundary between general conditions and fee is negotiated so carefully.

False. The honest primary reason is control of the critical path. When foundations and slab belong to your own superintendent, you can add a crew Thursday or place on a Saturday; when they belong to a subcontractor with three other jobs, you can only ask. Self-perform is a way to take a productivity risk you believe you can manage better than the market prices it — on Northgate's slab on grade, $66,680 of it, which is yours to win or lose.

False — and this is a clause you must write. A properly drafted subcontract includes an order-of-precedence provision making the subcontract and its scope exhibit govern over the subcontractor's proposal, plus a statement that qualifications and exclusions in the proposal are not part of the agreement unless expressly restated in an accepted-qualifications exhibit. Without that language, a sub can attach two pages with fourteen exclusions and argue the exclusions came with the price.

False. It creates an unbilled liability at the end: you collected for value you did not create, so the last 10% of the work has 4% of the money behind it and you finish the punch list on your own dime. It is also routinely rejected by architects, it fails the contractual requirement to allocate the contract sum to the portions of the work, and when a job goes bad it becomes the owner's first exhibit that your billing was never trustworthy.

False. As-built drafting, operation and maintenance manual compilation, warranty administration, and owner training coordination are Division 01 requirements and belong in general conditions — $58,000 on Northgate. It is the line rookies forget most reliably, because it is spent after the interesting part is over.

1. Estimate the trade yourself in detail from quantity and unit cost, and use your number as the check on theirs. An 8% gap is a negotiating position; a 40% gap is a scope misunderstanding or a market problem.
2. Break the package up — by area or by system — and see whether smaller firms come to the table. One elevator bidder is a fact of life; one drywall bidder means your package is too big or your terms scared people off.
3. Carry a qualified allowance and disclose it in the basis of estimate, with the exposure named.

What makes "carry it and hope" different: all three responses produce a documented position that somebody can evaluate. An allowance disclosed in the basis of estimate is a managed risk. An allowance hidden inside a division total is a lie with a timer on it. And in every case, ask why nobody else bid — the answer is usually available for the price of one phone call.

The rankings inverted because the base bids were priced against three different scopes. Cordova excluded shaft wall, firestopping, impact-resistant board, hoisting, cleanup, temporary protection, and patching — $411,000 of real work — and priced the remainder honestly. Bright Line included nearly everything and looked expensive. Leveling adds back what each bidder excluded so the comparison is between prices rather than between paragraphs, and once that is done Cordova is $196,822 above Bright Line.

What it proved about Kestrel's estimate: the in-house budget was $2,760,000, built from quantities and unit costs without seeing a single quote, and the true leveled market price was $2,778,000 — a variance of 0.65%. That is what a good detailed estimate is for. It does not replace the market. It tells you which quote to believe.

1. Estimate at the level of detail you intend to control at. A cost code nobody will accurately charge against produces confident garbage, which is worse than no data. If nobody will separate "partitions, level 2" from "partitions, level 3" on a Friday afternoon timecard, do not create the code.
2. Keep self-perform work coded by labor, material, and equipment separately. A subcontract can be one line — it is a fixed price you either owe or do not. Self-perform is where you find out whether you are making money, and you cannot find that out from a lump sum.
3. Make the schedule of values and the estimate reconcilable. Not identical — the SOV is negotiated and organized for progress measurement, so it gets subdivided by area and floor — but you must be able to walk any SOV line back to the estimate lines behind it.

All three are pre-award decisions because the structure of the estimate is the structure of the control system. Restructuring after award means re-coding invoices and timecards that have already been posted, which nobody ever finishes.

Order matters — check the things that are errors before you touch the things that are judgments.

First, the duration mismatch, because it is a fact and not an opinion (T−80 to T−70). Your general conditions are built on 480 days and the bid form says 450. Thirty calendar days at your burn rate is real money in your favor: if the burn rate is roughly $3,400/CD, that is about $102,000 to remove — legitimately, with a name and a mechanism. But do not remove it until you have asked the harder question: can the job actually be built in 450 days? If your schedule says 480 and the bid form says 450, you are not saving $102,000 — you are buying 30 days of liquidated-damage exposure. Check the schedule first. If 450 is achievable, take the $102,000. If it is not, you have discovered that this bid needs an acceleration line, not a savings line.

Second, the plugs (T−70 to T−55). $1,815,000 of plugs is 9.1% of the cost of work — far too much this close to the deadline, and earthwork at $1,140,000 is the dangerous one because earthwork has the widest bid spread of any trade. Call for numbers. Any plug replaced by a real leveled quote is a legitimate move in either direction. Any plug that survives to T−45 gets a risk load and gets announced out loud with its total.

Third, the spreadsheet controls (T−55 to T−50). Sum the columns two ways — by division and by package — and confirm they agree. Recompute bond and insurance on the current contract value, not the value from this morning. Print it and check it on paper. Run the $/SF check against comparable recreation centers.

Fourth, "what did we forget" (T−50 to T−45). Everybody speaks once.

What you tell the executive at T−45: the truth, with the parts separated.

"We're at $19,840,000. I can take $102,000 out of general conditions because the bid form is 450 days and we priced 480 — but only if Margo confirms we can build it in 450, and she hasn't yet. That's a real saving with a name on it. Earthwork and elevators are still plugs — $1,815,000, or 9.1% of the cost of work, all our own numbers with no competitive check. Getting from $19,840,000 to $19,500,000 means finding $340,000. After the general-conditions adjustment, I have $238,000 left to find, and right now I don't have a line I can point to and tell you why it should be lower. If you want to take it out anyway, that's your call to make and I'll write it down as a fee decision — but understand that we'd be doing it with 9% of the job still plugged."

That last sentence is the whole job. You have separated finding savings (named, mechanical, defensible) from buying the job (a judgment about market and appetite that belongs to someone with the authority to lose the company's money) — and you have made sure the decision, whichever way it goes, exists in writing.