Case Study 8-1 — Footing F-112

What a differing site condition actually costs, who pays for it, and why the strongest argument in the room was the one nobody on the contractor's side wanted to make.

All parties, projects, and figures here are Tier-3 illustrative composites.

Setup

The project. The Northgate Outpatient Pavilion — 132,000 gross square feet, four stories, structural steel frame on spread footings, for Meridian Health System. Kestrel Construction Group is the construction manager at risk under a $47,500,000 guaranteed maximum price and 565 calendar days. Liquidated damages are $5,500 per calendar day; the contractually agreed extended general-conditions rate is $5,150 per calendar day. Substantial completion is September 18, Year 2.

The people. Ray Alvarez, senior project manager. Dani Okonkwo, field engineer. Jamal Foster, self-perform concrete superintendent. Margo Deacon, general superintendent. Priyanka "Pri" Sethi, Meridian's owner's representative. Ivy Kowalczyk, field representative for Vantage Geotechnical, the geotechnical engineer of record. Ruth Caldwell, principal at Caldwell Structural, the structural engineer of record.

The constraint. The foundation package runs from May 10 to a scheduled completion of July 3, Year 1, and it is on the critical path. Steel erection is scheduled to start August 4, Year 1 — a date Kestrel is already nervous about for unrelated reasons involving an anchor-bolt submittal, which is a different story in a later chapter.

The document. The Vantage geotechnical report recommends spread footings bearing on medium-dense sand at an allowable bearing pressure of 4,000 psf. Fifteen borings were drilled across the 6.2-acre site. The boring location plan puts B-9 in the northeast quadrant, and the log for B-9 records standard penetration test blow counts of 4, 5, and 6 through a three-foot layer at roughly bottom-of-footing elevation.

Nobody on Kestrel's team had read the logs.

What Happens

Tuesday, 7:52 a.m.

Ivy Kowalczyk stops the pour at footing F-112 — a 10'-6" × 10'-6" × 36" footing under a heavily loaded interior column — after a hand-penetrometer check finds soft, gray, organic-smelling silt where the report calls for medium-dense sand.

Jamal keeps pouring the seven footings Ivy has already accepted and holds the balance. One loaded ready-mix truck is turned back for a short-load charge; a second cannot be turned and is dumped as waste. The pump truck and a nine-person crew stand down at 9:20 and are released at noon.

By 9:20, Ivy has walked the quadrant and flagged nine footings as suspect.

Tuesday, 11:30 a.m. — the notice

Ray sends Pri Sethi a written notice of a differing site condition. It is one page. It contains the date, the time of discovery, the location by footing mark, a factual description of the material encountered, the fact that work is stopped, the request that Meridian direct its geotechnical engineer to evaluate, and a reservation of rights as to cost and time. It contains no dollar figure and no accusation.

Ray: "Pri, this is a notice, not a claim. I don't know what it costs yet. I'm telling you today because if I tell you in three weeks, my contract says I've waived it, and neither of us wants to spend the fall arguing about that instead of about the actual money."

Pri: "Understood. Send Vantage. What does this do to July third?"

Ray: "I'll know Thursday."

Wednesday–Thursday — the re-evaluation

Vantage mobilizes a portable rig and hand augers, takes four additional borings in the northeast quadrant, and issues a revised recommendation Thursday afternoon: over-excavate 4 feet below bottom of footing at the nine affected locations, extend the over-excavation laterally on a 1:1 projection from the footing edge, and replace with imported engineered fill compacted to 98% of modified Proctor, tested every lift.

Dani computes the volume. Each pit measures roughly 18'-6" square (10'-6" footing plus 4 feet on each side at the base) by 4 feet deep:

18.5 × 18.5 × 4.0 = 1,369 CF ÷ 27 = 50.7 CY per footing, call it 51 CY

9 footings × 51 CY = 459 CY of over-excavation and replacement

The cost

The time

Wei Chen, project controls, runs a time-impact analysis against the current schedule update. The foundation package loses four work days in the northeast quadrant. Two of those days are absorbed by float in the foundation-wall sequence. Three calendar days flow through to the project completion date.

3 CD × $5,150/CD extended general conditions = $15,450

Total exposure: $59,200 + $15,450 = $74,650

The Entitlement Analysis

This is where the case gets interesting, and where Ray does the least comfortable thing in the file.

The Type I argument. A Type I differing site condition is a subsurface or latent physical condition differing materially from the conditions indicated in the contract documents. Kestrel's argument: the contract documents, through the geotechnical recommendations Caldwell designed to, indicated medium-dense sand at 4,000 psf. What was encountered was soft organic silt. That is a material difference.

The problem with the Type I argument. Boring log B-9 — a document furnished with the report — recorded N-values of 4, 5, and 6 in that quadrant at that elevation. The condition was disclosed, in the factual data, to any contractor who read it. A reasonable contractor performing a reasonable pre-bid review of the furnished data had notice that the northeast quadrant might not perform like the rest of the site.

The Type II argument. A Type II condition is an unknown physical condition of an unusual nature differing materially from those ordinarily encountered in work of this character. Soft pockets in an alluvial site are not unusual. This argument is weaker still.

Ray, to Dani, Thursday evening: "Write it up honestly. If we lead with 'we were blindsided' and they put B-9 on the table, we lose the room and we lose the next three change orders too. Lead with what's true: the recommendations governed the design, we relied on them, the pocket is bigger and worse than one boring suggested, and we're not asking Meridian to pay for the part we should have caught."

Dani: "That costs us money."

Ray: "It costs us less than being the contractor whose numbers you have to check."

What Kestrel actually submitted. A change-order request quantifying the full $74,650, with the four new Vantage borings attached showing the pocket extends well beyond what B-9 alone would have predicted, and an explicit acknowledgment that B-9 disclosed soft material in the quadrant.

The settlement

The residual is the part attributable to what B-9 disclosed. Because Kestrel is at risk under a GMP with a 75% owner / 25% Kestrel split of unused contingency, drawing $12,900 from the $1,320,000 construction contingency costs Kestrel 25% of $12,900 = $3,225 in forgone shared savings at the end of the job — not the full $12,900.

💰 Money check. The event cost $74,650. Meridian paid $61,750. Kestrel's true, unrecovered cost was $3,225 in shared savings, plus about eleven hours of Ray's and Dani's time. The reason it was that cheap is a one-page notice written at 11:30 on the day of discovery. The reason it was not free is a boring log nobody read.

Analysis

What worked.

1. Notice on the day of discovery. Not a claim, not a number — facts, location, time, and the contract article. Ray gave notice before he knew what it cost, which is the only time notice is ever actually available to you.
2. Stopping the pour. Jamal did not "pour it and sort it out." Concrete on unsuitable bearing is an unrecoverable defect that would have become a demolition and a structural repair, at ten times the cost.
3. Keeping the seven good footings going. Stopping the affected work is not the same as stopping the job. That decision saved a full crew-day.
4. Quantifying time separately from cost. A time-impact analysis run against the current schedule update, before the negotiation, is worth far more than a retroactive reconstruction. That distinction becomes the entire subject of Chapter 33.
5. Honest entitlement. Kestrel conceded the part it should have caught, and got the rest paid without a fight. Meridian's board approved a change order in eleven days rather than eleven weeks.

What failed, and the mechanism.

The failure is not "we hit bad soil." Bad soil is a normal condition of the work. The failure is that the factual data disclosing the risk was in Kestrel's own file for four months and nobody read it. The mechanism is ordinary and universal: the geotechnical report is not a contract drawing, it is not in the specification book, it does not come through the submittal log, and no one on a project team is formally assigned to read it. It lands in a folder during preconstruction and stays there.

Notice what the two-hour fix would have been. Overlay the boring location plan on the foundation plan at the same scale. Read each log at bottom-of-footing elevation. Flag the low ones. Measure the distance to the nearest footings. Put those footings on the look-ahead as watch items, brief the geotechnical field representative, and carry a named over-excavation contingency line. Half a day of a field engineer's time in March would have converted a $74,650 surprise into a priced, owned, drawn-down reserve — which is exactly what Chapter 6 said contingency is for.

The theme. The project is built twice. The first build — the estimate, the risk register, the reading of the furnished data — determines the second. F-112 was not decided at 7:52 on a Tuesday in May. It was decided in a conference room in February, when a folder went unopened.

Discussion Questions

1. Ray conceded part of the entitlement rather than pressing the strongest available argument. Was that the right call? Argue both sides, and identify what he was actually buying with the $12,900.
2. Kestrel drew the residual from construction contingency rather than absorbing it against fee. Under a GMP with a 75/25 savings split, what is the real cost of that decision, and to whom? Would your answer change if the contingency were already 60% consumed in month five of an 18-month job?
3. Suppose the contract had contained no differing-site-conditions clause at all. How would the conversation with Pri Sethi have gone, and what would Kestrel have carried in its GMP if it had known that at bid time?
4. Jamal poured the seven accepted footings while stopping the nine suspect ones. What could have gone wrong with that decision, and what would you have required in writing before allowing it?
5. Vantage was engaged by Meridian, not by Kestrel. How does that relationship shape what Ivy Kowalczyk can and cannot say to Jamal in the field, and what does it mean for how you document a field direction from a consultant who is not your consultant?

Your Turn

Take a real geotechnical report — public agency bid packages routinely publish them — or use the one from your own project. Spend ninety minutes doing exactly what Dani should have done in February:

1. Print the boring location plan and the foundation plan at matching scale and overlay them.
2. For each boring, write the bottom-of-footing elevation beside it and read the log across at that elevation.
3. Build a one-page watch list: boring, N-values, nearest footing marks, distance, and the reason you flagged it.
4. Write a two-sentence contingency justification for a named over-excavation line item, with a quantity and a unit price.

Then estimate what your watch list would be worth if you were wrong about it — and what it would be worth if you were right.