What Makes an Engineering Investigation Defensible in Court

Why the Quality of an Engineering Investigation Decides the Case

I've been retained more than once to review another engineer's investigation after opposing counsel poked a hole in it big enough to drive a forklift through. The report looked fine on the surface — bound, stamped, full of photos. The problem wasn't the conclusion. It was that the conclusion wasn't supported by anything a court or an adjuster could independently verify.

An engineering investigation isn't just a visit to a site with a camera and a clipboard. It's a documented, repeatable process that has to survive scrutiny from someone whose entire job is to find its weak point. If the investigation can't answer basic questions — what was measured, when, with what instrument, under what conditions — it doesn't matter how confident the final paragraph sounds. That report becomes a liability instead of an asset the moment it's deposed.

What a Defensible Field Investigation Actually Documents

A field investigation that will hold up needs to do more than describe damage in adjectives. It needs numbers tied to locations, dates, and methods. For a concrete cracking claim, that means crack width measured with a comparator or crack gauge at multiple points along the crack length — not just 'hairline' or 'significant' — recorded to the nearest 0.005 inch per ACI 224R guidance on acceptable crack widths. For suspected rebar corrosion, it means half-cell potential readings per ASTM C876, not a visual guess based on rust staining.

Photography matters more than most non-engineers realize. A photo without a scale reference, a compass orientation, or a tie to a floor plan grid is nearly useless in a dispute six months later when nobody remembers which stairwell it was taken in. Every photo in my field files gets logged with GPS coordinates or grid reference, direction of view, and a running photo log cross-referenced to the narrative report. When I hand that file to an attorney preparing for litigation expert witness work, every photo has to be traceable back to a specific finding.

Instrument calibration records, weather conditions at time of inspection, and a clear description of what was and wasn't accessible also belong in the file. If a roof could not be accessed because of standing water hazards, that limitation needs to be stated — not glossed over. Omissions discovered later by opposing experts are far more damaging than limitations disclosed up front.

Five Red Flags That Signal a Weak Engineering Investigation

When I'm asked to peer-review a prior investigation — which happens on roughly a third of my litigation files — I look for the same handful of problems every time.

• No dimensioned sketch or plan showing exactly where measurements or samples were taken relative to the building.
• Conclusions about cause (settlement, construction defect, storm damage) stated without any discussion of alternative causes that were considered and ruled out.
• Crack width, displacement, or deflection described qualitatively ('moderate,' 'severe') with no numeric data.
• Single-visit investigations on conditions that are inherently progressive, like differential settlement or active water intrusion, with no monitoring period.
• Photographs that can't be matched to a specific location in the narrative text.

Case File: A Parking Garage Facade Failure That Hinged on Missing Data

A regional carrier retained us after a precast concrete spandrel panel detached from the third level of a parking structure and fell onto a row of parked vehicles. No injuries, but the property owner's prior engineer had already issued a report attributing the failure to 'age-related deterioration' and recommending a full facade replacement at roughly $2.1 million. The carrier wanted a second opinion before approving that scope.

The original report had photos of spalled concrete and exposed rebar, but no half-cell potential survey, no chloride content testing, and no record of the embedded steel's actual cover depth. We went back and took 14 core samples across four elevations. Carbonation depth averaged 22mm against a design cover of 38mm at the failed panel — meaning the carbonation front hadn't actually reached the reinforcement at the failure location. Chloride content tests, however, showed 3.8 lbs/yd³ near curb lines exposed to deicing salt runoff, more than triple the ACI 318 corrosion threshold of roughly 1.0–1.3 lbs/yd³ for coated reinforcement in that exposure class.

The real failure mechanism was a corroded embedded connection angle at one specific panel tie, not generalized age-related concrete deterioration across the whole structure. That distinction cut the justified repair scope from full facade replacement to targeted connection repair and localized panel re-anchoring at nine locations — about $340,000 instead of $2.1 million. The carrier's exposure dropped by roughly $1.76 million because the second investigation produced data the first one never collected. This is the kind of detail covered in our broader look at parking garage failure mechanisms, and it's a pattern we see often enough that it shapes how we scope every structural inspection from day one.

Chain of Custody: Why Evidence Handling Gets Challenged in Depositions

Physical samples — concrete cores, failed fasteners, corroded rebar sections, roofing membrane cuttings — are evidence, and they get treated like evidence. Every sample needs a unique ID, a documented extraction location and date, a record of who took custody of it and when, and a description of how it was stored and transported to the testing lab. I've seen cross-examinations spend twenty minutes on whether a core sample sat in a hot truck bed for three days before testing, because that detail can plausibly affect compressive strength results.

This matters just as much on the insurance side as it does in litigation. An adjuster relying on lab results to validate or deny a six-figure claim needs confidence that the sample tested is actually the sample taken from the damaged structure, not a substitute or a sample that degraded in transit. Chain-of-custody gaps are one of the most common reasons an otherwise sound technical finding gets excluded or discounted.

How Insurance Adjusters Should Use Investigation Findings

An adjuster reading an engineering investigation report isn't expected to independently verify ACI crack-width tables or carbonation chemistry. But a few practical checks go a long way toward catching a weak report before it drives a claim decision. Does the report distinguish between pre-existing conditions and damage from the loss event in question? Does it include a clear timeline tying observed damage to a specific date of loss, especially for claims involving hurricane and storm damage where wind versus flood causation determines coverage? Is there a cost estimate tied directly to the documented scope of damage, or does the repair recommendation exceed what the findings actually support?

We've built our insurance claim support process around giving adjusters a report they can defend internally and, if needed, in coverage litigation — numeric findings, photo-documented locations, and a causation analysis that rules out alternative explanations rather than just asserting one.

What Attorneys Should Demand Before Trial or Settlement

Before relying on any engineering investigation in a demand letter, mediation, or trial exhibit, request the complete field file — not just the final report. That includes raw measurement data, the photo log with location references, lab results with chain-of-custody documentation, and any field notes. If the expert can't produce a complete file, or if the file doesn't match the narrative in the report, that's a problem you want to find before opposing counsel does.

It's also worth asking what alternative causes were investigated and ruled out, and why. A report that only supports one theory of causation, with no discussion of what else was considered, reads as advocacy rather than analysis to a judge who has seen a hundred expert reports. Our attorney resources section walks through the specific questions worth raising with any retained or opposing expert before deposition.

Working With a Firm That Builds Investigations to Withstand Challenge

Every investigation we open gets scoped as if it will end up in front of a judge, even when it's a straightforward insurance claim. That means instrumented measurements instead of adjectives, documented sampling protocols, and a written discussion of alternative causes even when the primary cause seems obvious on day one. It costs a little more time up front. It costs far less than a report that collapses under cross-examination or gets excluded under a Daubert challenge.

If you're evaluating a prior investigation, scoping a new one, or trying to decide whether a claim or case needs an independent technical review, our engineering risk assessment and independent engineering review services are built specifically for that gap — before the file gets built the wrong way, not after.

Frequently Asked Questions

What's the difference between an engineering inspection and an engineering investigation?

An inspection is typically a visual review against a standard — checking for code compliance or general condition. An investigation is a focused technical effort to determine the cause of a specific failure or defect, usually involving measurements, sampling, and causation analysis that can be defended under cross-examination.

How long should a thorough engineering investigation take?

It depends on the failure type and whether monitoring is needed. A straightforward crack or water intrusion case might involve one field visit of several hours plus lab turnaround of two to three weeks. Progressive conditions like settlement or active corrosion often require a monitoring period of 30 to 90 days to establish whether movement is ongoing before a final causation opinion is issued.

Can an insurance company challenge the findings of a property owner's engineering report?

Yes, and it happens regularly. If the original report lacks instrumented data, chain-of-custody documentation, or discussion of alternative causes, a carrier's retained engineer can often produce a competing analysis that significantly narrows the scope and cost of a claim, as in the parking garage case described above.

What documents should I request from an expert before trusting their investigation findings?

Request the complete field file: raw measurement data, the full photo log with location references, laboratory test results with chain-of-custody records, instrument calibration records, and notes on any alternative causes that were considered and ruled out. If the final report can't be traced back to this underlying data, treat the conclusions with caution.

Does a weak initial investigation mean a claim or case is lost?

No. A second, more rigorous investigation can often correct the record, as it did in the parking garage facade case where additional sampling reduced an initial $2.1 million repair estimate to roughly $340,000. The key is getting an independent technical review before a settlement or repair scope is finalized, not after.