Every geotechnical investigation begins with an admission it can never fully escape: the engineer samples a handful of points and infers the whole. A boring log describes a column of soil a few inches wide; the foundation it informs may span acres. Between those borings lies interpolation — educated, disciplined, defensible interpolation, but interpolation nonetheless. The ground does not read the report. When it behaves differently than the subsurface investigation predicted, the surprise does not stay in the field — it travels into change orders, schedule claims, and, often enough, into a lawsuit that names the geotechnical firm.

This is why subsurface work carries severity out of all proportion to its fee. A soils report may represent a fraction of a project’s design budget, yet the excavation, dewatering, and structural consequences that flow from it can run into the millions. Ames & Gough’s 2026 survey found that 82 percent of A/E professional liability insurers paid multimillion-dollar claims in 2025, with 60 percent reporting higher claim severity year over year and civil engineering ranking among the highest-risk disciplines. The geotechnical report is small on the invoice and large on the docket.

The contract language that governs “differing site conditions” is where that exposure is either contained or amplified. Federal practice — and most private contracts that borrow from it — recognizes two categories. A Type I condition differs materially from what the contract documents indicated. A Type II condition is unknown and unusual, differing materially from what would ordinarily be encountered in work of that character. Both shift cost, and both invite a familiar question after the fact: did the geotechnical report indicate the condition, disclaim it, or stay silent? The answer decides who pays.

Here is the uncomfortable truth about the report itself. The instrument the engineer produces to reduce uncertainty becomes, in litigation, the primary exhibit against them. Plaintiffs’ counsel reads it not as guidance but as a warranty — every recommendation, every characterization, every line that could be construed as a promise about conditions is parsed for what it did or did not say. The Geoprofessional Business Association has built much of its risk guidance around this reality, urging clear statements of reliance, purpose, and limitation so that a report meant for one client and one use is not stretched to cover parties and purposes it was never scoped for.

Two tools do the heavy lifting. The first is a geotechnical baseline report — long standard on tunneling and underground work, now spreading to other subsurface projects. A GBR states, in measurable and contractual terms, the conditions the contractor is entitled to assume. Conditions better than the baseline are the contractor’s gain; conditions materially worse become the owner’s risk, triggering compensation rather than dispute. It converts an argument about geology into arithmetic about a number everyone agreed to in advance. The second tool is a limitation of liability tied to fee — a clause capping the geotechnical firm’s exposure at the value of its services or a negotiated figure. It does not erase responsibility; it right-sizes it to what the engagement was actually worth.

And here is where contract language quietly creates uninsured exposure. A professional liability policy responds to your negligence — to a failure to meet the standard of care. It does not, as a rule, respond to liabilities you assumed by contract that you would not have owed under common law. When a firm signs an indemnity that guarantees subsurface conditions, warrants the accuracy of interpolated data, or accepts responsibility for a contractor’s means and methods, it may be trading an insured risk for an uninsured one. The signature feels routine. The gap it opens is not. The International Bar Association’s commentary on ground conditions and design liability, and ASCE’s own tracking of expanding liability for design professionals, point the same direction: courts are increasingly willing to enforce what the parties wrote, which means what you write matters more, not less.

The discipline, then, is unglamorous and decisive. Scope the report to a defined client and purpose. Document the assumptions and the limits of the investigation. Say plainly what the data supports and what it does not. Negotiate the differing-site-conditions clause, the baseline, and the liability cap before the drilling rig arrives — not after the surprise. The firm that treats its contract as an afterthought inherits risks its policy was never built to carry; the firm that treats it as an engineering document controls them.

That is the work we do at PFTN. Our 4-Step Strategic Process — Strategic Discovery, Risk Assessment, Solution Design, and Ongoing Optimization — begins by illuminating the exact seams between what your contracts promise and what your policy will actually pay, then closes them intentionally, before a boring log becomes an exhibit. The uncertainty in the ground is permanent. The uncertainty in your risk transfer does not have to be.

— Ryan Mefford, President & Risk Advisor