Checklist · Tort

Defending a Products Liability Claim: A Practical Checklist

Type: Checklist
Jurisdiction: National
Date: 2024-08-01
Topic: Tort

A products liability complaint is not one case but three or four cases stapled together, and the defense that wins is the one that separates them early. This checklist walks the defense of a products claim from the first hour — preserving the subject product, the exemplar, and the entire design record — through joint inspection protocols, the design history file, chain-of-distribution and innocent-seller challenges, theory-by-theory attack on manufacturing, design, and warning defects, state-of-the-art and compliance evidence, misuse and alteration defenses, an early preemption analysis that can end the case outright, Daubert motions against the plaintiff's design and causation experts, the fight over other-similar-incident evidence, recall and post-sale duty exposure, insurance tender and supply-chain indemnity, and the punitive-damages narrative. Products law is overwhelmingly STATE law: the governing design-defect test, the availability of state-of-the-art and sealed-container defenses, and the treatment of comparative fault all vary, so confirm the forum's rule before you build the file. Every phase closes with the traps that turn a defensible product into an indefensible defendant.

Defending a Products Liability Claim: A Practical Checklist

Tort > Products Liability | Published 1 August 2024 - Updated 14 December 2025 | UpLaw - uplaw.us

Most products cases are not lost at trial. They are lost in the first ninety days — when the subject product is returned to the customer, the design engineer's inbox auto-deletes, and a well-meaning quality manager writes an email that becomes the plaintiff's opening slide.

Keywords: products liability defense, design defect, failure to warn, reasonable alternative design, risk-utility, consumer expectations, Daubert, Rule 702, federal preemption, learned intermediary, state of the art, product misuse, spoliation, other similar incidents, punitive damages

How to use this checklist

A products liability complaint typically pleads strict liability, negligence, and breach of warranty, and within strict liability it pleads three distinct defect theories — manufacturing, design, and warning. Each of those has its own elements, its own proof, its own defenses, and its own path to summary judgment. The single most common defense error is treating the case as one undifferentiated "product case" and building one undifferentiated defense. It is not. It is three cases, and you may win two of them on the papers.

Work the phases below in order. Phases 1 through 3 are clock-driven and start the day you learn of the incident — often before suit. Phases 4 through 9 are the substantive defense build. Phases 10 through 15 are the dispositive and exposure-management moves. For the doctrinal underpinnings of everything here, work alongside Products Liability: A Comprehensive Guide to Design Defect, Manufacturing Defect, and Failure to Warn and Tort Law: A Comprehensive Guide to Negligence, Intentional Torts, and Strict Liability.

A running example. Alder & Vane Machine Works builds the AV-720 benchtop table saw. A cabinetmaker, injured when his hand contacted the blade during a rip cut, sues in state court alleging that the AV-720 was defectively designed because it lacked an active injury-mitigation system (a flesh-detection brake), that the riving knife and blade guard were inadequate, and that the warnings failed to convey the kickback hazard. He also alleges the specific saw's guard-mounting bracket was mis-welded. Alder & Vane will follow this checklist.

Phase 1 — Preserve everything: the subject product, the exemplar, and the design record

[ ] Issue a written litigation hold the day you have notice of a claim or a reasonably anticipated claim — not the day suit is filed.
[ ] Scope the hold to custodians and systems: design engineering, quality, manufacturing, regulatory, customer service, warranty, sales, marketing, and the CAD/PLM system.
[ ] Suspend auto-deletion on email, chat (Slack/Teams), ticketing systems, and the warranty/complaint database.
[ ] Identify and physically sequester the subject product — the actual unit involved — and every piece of it, including debris, packaging, and the manual.
[ ] Identify and sequester an exemplar: a same-model, same-vintage unit, unaltered, that can be tested and shown to a jury.
[ ] Preserve the manufacturing record for the subject unit's lot: build records, inspection results, torque logs, supplier certificates of conformance.
[ ] Preserve prototypes, failed design iterations, and test rigs — do not "clean up the lab."
[ ] Instruct field and service personnel in writing: do not repair, disassemble, discard, or "make safe" the subject unit.
[ ] Document the chain of custody from the moment the unit enters your control.

Why it matters, and the traps. Fed. R. Civ. P. 37(e) governs the loss of electronically stored information and permits severe sanctions — including an adverse-inference instruction or dismissal — only on a finding that the party "acted with the intent to deprive." That intent requirement is a real protection, but it applies to ESI. The loss of tangible evidence — the saw itself — is governed by the forum's spoliation law and the court's inherent power, and many state courts will give an adverse-inference instruction on a far lower showing, sometimes mere negligence. A manufacturer that lets a routine field-service policy send the subject saw back into a customer's shop has handed the plaintiff a spoliation instruction that no expert can undo.

The trap that recurs: the hold is issued to Legal and the C-suite but not to the shop floor. Alder & Vane's regional service technician, following a twenty-year-old policy, "reconditioned" three returned AV-720 units the week the demand letter arrived, destroying the best comparator population. Work through Issuing a Litigation Hold and Preserving ESI: A Practical Checklist and extend it, expressly, to physical evidence.

Phase 2 — Secure and inspect the actual unit; negotiate a joint inspection protocol

[ ] Determine who has the subject product: plaintiff, plaintiff's counsel, a public agency, an employer, a subrogated insurer, or a salvage yard.
[ ] If someone else holds it, send an immediate preservation demand and, if necessary, move for an inspection order.
[ ] Negotiate a written joint inspection protocol before anyone touches the unit.
[ ] The protocol should specify: who attends, what may be photographed, what may be measured, what (if anything) may be disassembled, the sequence of disassembly, who performs it, and what destructive testing requires further agreement or court order.
[ ] Photograph and 3D-scan the as-received condition before any disassembly.
[ ] Preserve the "as-found" configuration: guard position, fence setting, blade installed, aftermarket parts, missing fasteners.
[ ] Reserve rights: nothing in the protocol concedes authenticity, condition, or relevance.
[ ] Run parallel non-destructive testing on the exemplar so the two can be compared.

Why it matters, and the traps. The condition of the subject product at the time of the accident is the single most contested fact in most products cases, and it is usually resolved by whoever documented it first and best. Non-destructive imaging before disassembly is cheap; the ability to prove that the guard was removed before the accident, and that the removal was done by a third party with a wrench, is priceless.

The trap: agreeing to destructive testing without a protocol, then discovering that the plaintiff's metallurgist sectioned the guard-mounting bracket along exactly the plane that would have shown the weld was intact. In the AV-720 case, the joint protocol required that any sectioning be done by a neutral lab, with both experts present and the full weld cross-section preserved. That single paragraph is what allowed Alder & Vane to defeat the manufacturing-defect theory outright.

Phase 3 — Assemble the design history, testing, and complaint record (assume it all gets produced)

[ ] Collect the design history file: requirements documents, design reviews, FMEA (failure modes and effects analysis), hazard analyses, risk assessments.
[ ] Collect all validation and verification testing — including the tests the product failed before it passed.
[ ] Collect the standards-compliance record (UL, ANSI, ISO, OSHA-referenced consensus standards) and the certification files.
[ ] Pull the complaint, warranty, and field-incident history for the model line, and define the date range and product scope you are prepared to defend.
[ ] Pull marketing materials and the operator's manual in every revision, with revision dates.
[ ] Identify privileged material early and route it correctly — but do not assume an engineering risk assessment is privileged simply because a lawyer received a copy.
[ ] Map the custodians and systems for a defensible collection.
[ ] Prepare the corporate designee(s) early; assume a Fed. R. Civ. P. 30(b)(6) notice on design, testing, complaints, and warnings.

Why it matters, and the traps. Everything in an FMEA is written in the language of hazard and severity, because that is what an FMEA is for. A row reading "Hazard: blade contact. Severity: 9. Mitigation: guard + warning label" is good engineering; on a courtroom screen, blown up to eight feet, it reads as a confession. You cannot suppress it, and you should not want to — the defense of a well-run safety process is that the process found the hazard and addressed it. But you must know what is in the file before the plaintiff does, and your 30(b)(6) witness must be able to explain, calmly, what a severity score is and is not.

The traps are three. First, treating the risk file as privileged: ordinary-course engineering analyses are ordinary-course business records, and copying counsel does not transform them. Build a real privilege log; see Building a Privilege Log: A Practical Checklist. Second, under-scoping the complaint-history search and then having to supplement — which converts a document problem into a credibility problem. Third, letting the plaintiff define the production scope for you; respond to document requests with a considered scope objection rather than a reflexive one, per Responding to Rule 34 Document Requests: A Practical Checklist.

Phase 4 — Are you even the right defendant? Chain of distribution, components, and successors

[ ] Confirm you actually manufactured, sold, or distributed this unit — check serial number, build date, and plant.
[ ] Determine whether the unit is genuine or counterfeit, and whether it was sold into your authorized channel.
[ ] If you are a non-manufacturing seller or distributor, check the forum's innocent-seller / sealed-container statute (Texas Civ. Prac. & Rem. Code ch. 82 is a representative example; many states have analogues and many do not).
[ ] If you supplied a component, evaluate the component-parts doctrine: a supplier of a non-defective generic component is generally not liable for the integrator's design, unless it substantially participated in the integration.
[ ] Analyze successor liability if the product line was acquired: mere continuation, de facto merger, product-line continuity (a minority rule), and express assumption.
[ ] Check the statute of limitations and the statute of repose — repose periods for products and improvements to realty vary widely by state, and the federal General Aviation Revitalization Act of 1994 imposes an 18-year repose for general-aviation aircraft.
[ ] Preserve every one of these as an affirmative defense in the answer.

Why it matters, and the traps. Products plaintiffs name everyone in the chain and sort it out later. A defendant that can show it is the wrong party, or that a statute immunizes it, exits the case for the cost of a motion. But these defenses are waivable, and the statute of repose in particular is easy to miss because it runs from the date of sale or first use, not the date of injury — a distinction that surprises lawyers trained on limitations. Plead them all; see Answering a Complaint and Pleading Affirmative Defenses: A Practical Checklist.

The trap: assuming the component-parts doctrine covers you when your engineers sat in the customer's design reviews. Substantial participation in the integration is the doctrine's exception, and email is where it is proved.

Phase 5 — Separate the defect theories and attack each on its own terms

[ ] Chart the complaint: which paragraphs plead manufacturing defect, which plead design defect, which plead warning defect, and which plead negligence or warranty on the same facts.
[ ] Manufacturing defect (Restatement (Third) of Torts: Products Liability § 2(a)): the unit departed from the manufacturer's own specifications. Test the subject unit against spec. If it conforms, this theory dies.
[ ] Design defect (§ 2(b)): the entire product line is condemned. This is the theory with the recall-scale exposure.
[ ] Warning defect (§ 2(c)): foreseeable risks could have been reduced by reasonable instructions or warnings.
[ ] Confirm which theories survive the forum's economic-loss rule and which are barred as contract claims dressed as torts.
[ ] Serve contention discovery forcing the plaintiff to commit to a defect theory and to identify the defect with specificity.
[ ] Move for partial summary judgment theory by theory — do not wait for a single omnibus motion.

Why it matters, and the traps. A manufacturing-defect claim is narrow and fact-bound: the unit either met spec or it did not. A design-defect claim is an attack on every unit you have ever sold. Collapsing them lets the plaintiff try a manufacturing case and argue a design case, and lets him use "the weld was bad" to prove "the design was bad."

In the AV-720 case, the joint inspection established that the guard-mounting weld conformed to spec, with penetration within tolerance. Alder & Vane moved for partial summary judgment on the manufacturing-defect count alone and won it — which mattered far beyond the count itself, because it stripped the plaintiff of his most visceral fact. Use Serving Written Discovery Requests: A Practical Checklist to force the contention commitments, and Moving for Summary Judgment Under Rule 56: A Practical Checklist to convert them into judgments.

Phase 6 — Design defect: which test applies, and is a reasonable alternative design required?

[ ] Determine the forum's design-defect test. This varies by state and is the most important research task in the case.
[ ] Risk-utility / reasonable alternative design (RAD): the Restatement (Third) § 2(b) approach, adopted in many states — the plaintiff must prove a reasonable alternative design existed and its omission rendered the product not reasonably safe.
[ ] Consumer expectations: the older Restatement (Second) § 402A approach — the product is defective if it is more dangerous than an ordinary consumer would expect.
[ ] Hybrid / two-prong: California permits either prong depending on the product's complexity, with the burden shifting to the defendant on risk-utility (Barker v. Lull Engineering Co., 20 Cal. 3d 413 (1978)), and reserves consumer expectations for cases within ordinary experience (Soule v. General Motors Corp., 8 Cal. 4th 548 (1994)).
[ ] Note the outliers: some states (Connecticut, in Potter v. Chicago Pneumatic Tool Co. (Conn. 1997)) expressly decline to make a reasonable alternative design a required element in every case.
[ ] If RAD is required, attack it: was the alternative technologically feasible at the time of sale, economically practicable, and did it not introduce offsetting hazards?
[ ] Demand that the plaintiff's expert actually build, test, and quantify the alternative — a sketch is not a design.
[ ] Evaluate the design in the context of the whole product: an alternative that eliminates one hazard while creating another is not "reasonable."

Why it matters, and the traps. The design test is dispositive of how the case is tried. Under a strict RAD regime, a plaintiff who has not tested his alternative often cannot get to a jury. Under consumer expectations, a jury may condemn a design it does not understand. Never assume the forum's rule; verify it, and verify whether the forum limits consumer expectations to products within ordinary lay experience.

Alder & Vane's plaintiff proposed flesh-detection braking as the alternative design. The defense attacked it on the axis that matters: availability at the time of sale, licensing terms and cost as actually offered, the false-trip rate on wet or conductive stock, and the fact that the alternative would not have prevented the kickback mechanism the plaintiff's own biomechanics expert described. That is a RAD attack; "our saw was fine" is not.

Phase 7 — Warnings: adequacy, learned intermediary, sophisticated user, and the heeding presumption

[ ] Collect every warning: on-product labels, the manual, packaging, point-of-sale material, and training.
[ ] Assess adequacy on content, conspicuousness, and location — and against the applicable consensus standard (ANSI Z535 for safety signs is the common benchmark).
[ ] Attack causation: would a different warning have changed this plaintiff's conduct? Depose him on whether he read the manual, saw the label, or removed the guard.
[ ] Determine whether the forum applies a heeding presumption (that an adequate warning would have been followed) and whether it is rebuttable.
[ ] If the product reaches the user through a professional, invoke the learned intermediary doctrine — the warning runs to the prescribing physician, not the patient. Recognized in the large majority of states; a handful have rejected or limited it.
[ ] If the buyer is a knowledgeable commercial user, invoke the sophisticated user / bulk supplier defense.
[ ] Argue the open and obvious hazard where the forum recognizes it as bearing on duty or on adequacy.
[ ] For prescription drugs and devices, consider Restatement (Second) § 402A cmt. k (unavoidably unsafe products).

Why it matters, and the traps. Failure-to-warn is the theory that survives when the design theory fails, because it costs a plaintiff nothing to allege that a different label would have worked. It is also the theory most vulnerable to a causation attack — and the deposition is where you win it. A plaintiff who concedes he never opened the manual and had removed the guard on prior cuts has destroyed his own warning claim, whatever the label said. Prepare that examination deliberately; see Preparing to Take a Deposition: A Practical Checklist and The Federal Deposition Toolkit: From Notice to Trial Testimony.

The trap: over-warning. Plaintiffs argue that a label crowded with forty warnings dilutes the critical one. If your product carries a wall of text, be ready to explain the hierarchy — and note that the heeding presumption, where it exists, can shift the causation burden onto you.

Phase 8 — State of the art, industry standards, and regulatory compliance

[ ] Build the state-of-the-art record: what was technologically and economically feasible at the time the product was sold, not today.
[ ] Confirm the forum's rule. This varies: some states treat state of the art as a complete defense by statute; others treat it as evidence relevant to risk-utility; a few exclude it from strict-liability design claims altogether.
[ ] Assemble compliance evidence: applicable consensus standards, certification listings, and any federal regulatory standard the product meets.
[ ] Understand what compliance does: in most states it is evidence of due care but is not a complete defense, and non-compliance is often evidence of defect (an asymmetry worth acknowledging before the plaintiff does).
[ ] Check for a statutory compliance presumption — a number of states create a rebuttable presumption of non-defectiveness for products meeting mandatory federal standards.
[ ] Prepare for Fed. R. Evid. 407: your post-accident design change is inadmissible to prove defect or a need for a warning, but is admissible for impeachment or, if controverted, to show feasibility — so do not gratuitously controvert feasibility.
[ ] Distinguish standards you helped write; plaintiffs will argue industry capture.

Why it matters, and the traps. Compliance evidence is the backbone of the corporate-conduct defense, and the corporate-conduct defense is what keeps punitive damages out of the case. But compliance is a floor, not a ceiling, and juries are told so. The persuasive version is not "we met the standard" but "we met the standard, we tested beyond it, and here is the engineering judgment behind each choice."

The Rule 407 trap is self-inflicted. Alder & Vane redesigned the AV-720's riving knife eighteen months after the accident for unrelated manufacturability reasons. That change is inadmissible to prove the original design was defective. It becomes admissible the moment defense counsel argues, broadly, that "a better guard was not feasible." Discipline the feasibility argument; see The Federal Rules of Evidence: A Comprehensive Guide.

Phase 9 — Plaintiff-conduct defenses: misuse, alteration, comparative fault, assumption of risk

[ ] Reconstruct exactly how the product was being used at the moment of injury.
[ ] Misuse: was the use unforeseeable? Foreseeable misuse must be designed and warned against; unforeseeable misuse is a defense. Frame it correctly.
[ ] Substantial alteration / modification: was a guard removed, an interlock bypassed, a part substituted? Document who did it and when.
[ ] Check whether the forum makes substantial alteration a complete bar or merely a comparative-fault factor.
[ ] Comparative fault: determine the forum's regime — pure comparative, modified (50% or 51% bar), or, in a small minority (Alabama, Maryland, North Carolina, Virginia, and the District of Columbia), contributory negligence as a complete bar.
[ ] Determine whether the forum allows comparative fault to reduce a strict liability recovery at all — a genuine split.
[ ] Assumption of risk: available in some states; merged into comparative fault in others.
[ ] Identify empty-chair and third-party fault: the employer, the modifier, the trainer, the used-equipment reseller. Check the forum's allocation and joint-and-several liability rules.
[ ] Preserve workplace-injury facts: OSHA citations, employer training records, prior near-misses on the shop floor.

Why it matters, and the traps. These defenses are how a defensible design becomes an affordable verdict. But the misuse defense is routinely mispleaded. "He used it wrong" is not a defense; "he used it in a way no manufacturer could reasonably have anticipated" is. If your own manual anticipates the misuse, or your FMEA scored it, you have proved foreseeability against yourself.

Alder & Vane's plaintiff had removed the blade guard and the anti-kickback pawls and was free-hand ripping a warped board without the fence — a use the manual expressly prohibits. That is not a misuse defense so much as a causation and comparative-fault case, and it is stronger for being framed honestly. For the damages side of that allocation, see Evaluating a Personal Injury Claim: A Practical Checklist and The Litigation Damages Toolkit: Valuing and Proving Money Awards.

Phase 10 — Federal preemption: evaluate it EARLY, because it can end the case

[ ] Ask first: is the product regulated by a federal scheme with an express preemption clause?
[ ] Medical devices: a Class III device approved through premarket approval is protected by the Medical Device Amendments' preemption clause, 21 U.S.C. § 360k(a) — Riegel v. Medtronic, Inc., 552 U.S. 312 (2008). Devices cleared through the § 510(k) substantial-equivalence route are not so protected.
[ ] Brand-name prescription drugs: Wyeth v. Levine, 555 U.S. 555 (2009) — no impossibility preemption where the manufacturer could have strengthened the label unilaterally under the changes-being-effected regulation.
[ ] Generic drugs: PLIVA, Inc. v. Mensing, 564 U.S. 604 (2011) (failure-to-warn preempted by the duty of sameness) and Mutual Pharmaceutical Co. v. Bartlett, 570 U.S. 472 (2013) (design-defect claim preempted).
[ ] Motor vehicles: Geier v. American Honda Motor Co., 529 U.S. 861 (2000) (implied conflict preemption where the federal standard gave manufacturers a deliberate choice) — but compare Williamson v. Mazda Motor of America, 562 U.S. 323 (2011) (no preemption where the agency's choice was not a significant regulatory objective). The Safety Act's savings clause is 49 U.S.C. § 30103(b).
[ ] Pesticides: Bates v. Dow Agrosciences LLC, 544 U.S. 431 (2005), construing FIFRA's preemption clause, 7 U.S.C. § 136v(b) — state-law claims survive if they impose requirements parallel to the federal labeling requirements.
[ ] Government contracts: the government contractor defense, Boyle v. United Technologies Corp., 487 U.S. 500 (1988) — reasonably precise specifications approved by the United States, conformity, and warning of known dangers.
[ ] If preemption is available, plead it, brief it early, and consider removal and consolidation strategy around it.

Why it matters, and the traps. Preemption is the only defense in this checklist that can dispose of the entire case, including the punitive count, without a single expert. It is also the defense most often discovered in year two. Screen for it in the first thirty days.

The traps: assuming that any FDA involvement preempts (it does not — the § 510(k) pathway is regulatory clearance, not the "specific" federal requirement Riegel demanded); assuming a savings clause defeats implied conflict preemption (Geier holds it does not); and forgetting that a preemption defense must usually be pleaded to be preserved. Because preemption turns on the scope and force of agency action, the analysis increasingly borrows from ordinary administrative law — see Administrative Law: A Comprehensive Guide to Agency Action and Judicial Review.

One currency point matters here, and briefs still get it wrong. Preemption arguments used to lean on Chevron-style deference to an agency's reading of its own organic statute. Chevron deference no longer exists: it was overruled in Loper Bright Enterprises v. Raimondo, 603 U.S. 369 (2024), and courts now determine the best reading of a statute de novo, giving an agency's view only the weight its persuasiveness earns under Skidmore v. Swift & Co., 323 U.S. 134 (1944). Do not tell a court that it must defer to the agency's view of its preemptive reach. The good news for a defendant is that the preemption cases in this phase never rested on Chevron in the first place — Riegel, Geier, Mensing, and Bartlett turn on the statutory text, the impossibility of simultaneous compliance, and the conflict with a federal objective, and all of that survives Loper Bright untouched. Argue the statute and the conflict, not the deference.

Phase 11 — Daubert: move to exclude the plaintiff's design and causation experts

[ ] Read the report against Fed. R. Civ. P. 26(a)(2)(B): does it disclose all opinions, the basis, the facts considered, and the exhibits?
[ ] Apply Fed. R. Evid. 702 as amended effective December 1, 2023: the proponent must show by a preponderance of the evidence that the criteria are met, and that the opinion "reflects a reliable application" of the methodology to the facts.
[ ] Apply Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579 (1993): testability, peer review, error rate, standards, general acceptance — a flexible, non-exhaustive list.
[ ] Apply Kumho Tire Co. v. Carmichael, 526 U.S. 137 (1999): the gatekeeping obligation extends to engineering and other technical experts, not just scientists. This is the workhorse citation in a products case.
[ ] Apply General Electric Co. v. Joiner, 522 U.S. 136 (1997): a court may exclude an opinion when there is "simply too great an analytical gap" between the data and the opinion; the standard of review is abuse of discretion.
[ ] Target the design expert: did he build and test the alternative design? Quantify its risk reduction? Account for offsetting hazards? Or did he draw a picture?
[ ] Target the causation expert: general causation (can this defect cause this injury?) then specific causation (did it here?); differential diagnosis must rule in and rule out.
[ ] Target the warnings expert: is "human factors" here anything more than the ipse dixit that a bigger label works better?
[ ] Preserve the motion for trial and for appeal; get a ruling on the record.

Why it matters, and the traps. In a design case the plaintiff usually cannot reach a jury without an expert. Exclude the expert and, in most jurisdictions, summary judgment follows automatically. The 2023 amendment to Rule 702 was adopted precisely because courts had been admitting shaky opinions and calling reliability a question of weight; it makes explicit that reliability is an admissibility question the proponent must carry.

Alder & Vane's motion did not argue that flesh-detection braking is a bad idea. It argued that the plaintiff's expert never installed such a system on an AV-720, never tested it, never measured false-trip rates on the wet stock the plaintiff was cutting, and could not say the system would have engaged before the kickback drove the plaintiff's hand into the blade. That is a Joiner analytical gap and a Kumho methodology failure, and it is the shape of a winning motion. Build it with Challenging an Expert Under Daubert and Rule 702: A Practical Checklist.

Phase 12 — Other similar incidents and pattern evidence: fight the scope

[ ] Define the plaintiff's OSI demand precisely: which models, which years, which failure mode, which injury.
[ ] Insist on the forum's substantial similarity requirement — same product, same defect, same mechanism, and substantially similar circumstances.
[ ] Argue Fed. R. Evid. 401/403 (or the state analogue): dissimilar incidents are irrelevant and, if admitted, invite a mini-trial on each.
[ ] Distinguish the purpose: courts often relax similarity when OSI evidence is offered on notice, and tighten it when offered on defect or causation. Move to limit the purpose.
[ ] Seek a protective order and a phased production; resist a fishing expedition across an entire product family.
[ ] Audit your own complaint database before producing it — know your worst document.
[ ] Prepare rebuttal: exposure denominators (incidents per million operating hours) are the defense's best answer to a raw incident count.

Why it matters, and the traps. Nothing moves a products jury like a list of other injured people. Nothing distorts a products case like a list of dissimilar other injured people. The fight over the OSI universe is often the most consequential discovery fight in the case, and it should be litigated with a proposed order in hand, not resisted with a boilerplate objection. Where a claimed pattern is broad enough to attract class treatment, the same denominator work supports the predominance attack in Opposing Class Certification: A Defense Checklist and the framework in Class Action Litigation: A Comprehensive Guide to Rule 23.

Phase 13 — Recall exposure, post-sale duties, and CPSC obligations

[ ] Determine whether the incident triggers a reporting duty under the Consumer Product Safety Act, 15 U.S.C. § 2064(b) — a product that fails to comply with a safety rule, contains a defect that "could create a substantial product hazard," or creates an unreasonable risk of serious injury or death.
[ ] Note the timing: CPSC's regulations, 16 C.F.R. Part 1115, treat the obligation as immediate (the agency's stated benchmark is 24 hours once the firm has reportable information), with a short internal-evaluation window. Confirm the current rule and the agency's current guidance before you calendar anything — reporting timelines, civil penalty maxima (which are inflation-adjusted annually), and the agency's interpretive guidance are amended regularly.
[ ] Track the separate § 2084 duty to report certain settled or adjudicated lawsuits alleging death or grievous bodily injury.
[ ] Evaluate the post-sale duty to warn (Restatement (Third) § 10) and the duty to recall (§ 11) — the latter is much narrower and generally arises only from a statute, regulation, or a voluntarily undertaken recall performed negligently.
[ ] Coordinate the regulatory workstream with the litigation workstream — but do not let litigation posture delay a required report.
[ ] Understand the evidentiary posture: a § 15(b) report is not itself an admission of a defect, but plaintiffs will offer the surrounding facts, and a late report is punitive-damages fuel.
[ ] Run any corrective action through Managing a Product Recall: A Practical Checklist.

Why it matters, and the traps. The reflex to "wait until the litigation clarifies things" before reporting is the single most dangerous instinct in this area. Late reporting is independently sanctionable, it converts a defensible product case into a corporate-conduct case, and it hands the plaintiff a punitive theory that has nothing to do with the design. The correct sequencing is: evaluate the reporting duty on the regulatory clock, and build the litigation defense around whatever that evaluation requires.

Phase 14 — Insurance tender and indemnity up the supply chain

[ ] Tender to every potentially responsive policy: CGL, products/completed operations, excess and umbrella, and any product-recall policy.
[ ] Tender promptly and in writing; late notice is a coverage defense in many states.
[ ] Reserve rights against the insurer's reservation of rights; do not concede the duty-to-defend analysis.
[ ] Check whether you are an additional insured on a supplier's or customer's policy.
[ ] Review the supply agreements for indemnity, defense, and insurance-procurement clauses — and tender to the component supplier.
[ ] Evaluate contribution and indemnity (contractual, and common-law implied) against upstream suppliers and downstream modifiers.
[ ] Preserve the right to bring third-party claims within the forum's impleader deadline.
[ ] Watch the coordination problem: the insurer's cost-driven settlement instinct may collide with your interest in defending the design.

Why it matters, and the traps. The defense of a products case is often funded by someone else, and the strategic latitude you have depends entirely on how carefully the tender was made. Work the coverage question in parallel from day one using Tendering a Claim and Preserving Coverage: A Policyholder's Checklist and Insurance Coverage Law: A Comprehensive Guide for Policyholders.

The trap: settling a case cheaply on the insurer's advice when the design theory would have been vindicated — and thereby inviting the next fifty complaints. Products defense has a portfolio dimension that a single claim file does not capture.

Phase 15 — Punitive damages and the corporate-conduct narrative

[ ] Identify the forum's punitive standard: malice, fraud, conscious disregard, or "reckless indifference" — and its burden (often clear and convincing evidence).
[ ] Move to bifurcate liability from punitive damages and net worth where the forum permits.
[ ] Audit the documents that will be used to prove conscious disregard: cost-benefit memoranda quantifying lives against dollars, complaint trends that were tracked but not acted on, and the internal email that reads worse than the decision it describes.
[ ] Build the countervailing record: what the company did — the testing, the standard exceeded, the warning added, the design change made.
[ ] Invoke the constitutional guardrails: BMW of North America, Inc. v. Gore, 517 U.S. 559 (1996) (reprehensibility, ratio, comparable penalties); State Farm Mutual Automobile Insurance Co. v. Campbell, 538 U.S. 408 (2003) (single-digit ratios ordinarily the outer limit); Philip Morris USA v. Williams, 549 U.S. 346 (2007) (a jury may not punish for harm to non-parties).
[ ] Request limiting instructions on nationwide conduct and out-of-state harm.
[ ] Check the forum's punitive damages cap — many states have one, and they vary enormously.

Why it matters, and the traps. Products cases are compensatory on the pleadings and punitive on the evidence. The plaintiff's trial story is almost never "this saw was badly designed"; it is "this company knew and did nothing." Every phase above feeds this one. The trap is to defend fifteen phases of engineering and never build the affirmative narrative of a company that ran a safety process, found the hazard, and made a defensible choice. For the trial-level execution, see Federal Civil Trial Practice: From Final Pretrial Conference to Verdict.

Common mistakes to avoid

Issuing the hold to lawyers, not to the shop floor. Physical spoliation is judged by the forum's law, not by the intent standard of Fed. R. Civ. P. 37(e).
Letting the subject product be inspected without a written protocol. The disassembly sequence is the case.
Treating engineering risk analyses as privileged. FMEAs, hazard analyses, and design reviews are ordinary-course records. Know them; do not hide them.
Defending "the product" instead of defending each theory. You may win the manufacturing count on the papers and never think to move.
Assuming the forum's design-defect test. Consumer expectations, risk-utility, hybrid, and RAD-required regimes produce different cases. Research it before you retain the expert.
Pleading "misuse" as "he used it wrong." Foreseeable misuse is your problem, not his. Frame the defense as unforeseeability, or frame it as comparative fault.
Discovering preemption in year two. It is the only complete defense on this list. Screen for it in month one.
Broadly controverting feasibility and thereby opening the door under Fed. R. Evid. 407 to your own post-accident design change.
Producing an unbounded other-similar-incidents universe because the objection felt too hard to litigate.
Delaying a CPSA § 15(b) report for litigation reasons. The regulatory clock does not pause for the defense strategy, and lateness is punitive-damages evidence.
Late or partial insurance tender. Notice provisions are enforced.
Sending a corporate designee to a Rule 30(b)(6) deposition unprepared to explain the safety process. That witness is the company; a defensive, evasive one is worth more to the plaintiff than any document.

The consolidated defense checklist

First 30 days

[ ] Litigation hold issued — custodians, systems, and physical evidence
[ ] Subject product located and sequestered; chain of custody documented
[ ] Exemplar secured
[ ] Preservation demand sent to any third-party holder
[ ] Insurance tendered in writing to all layers
[ ] Preemption screen completed
[ ] Statute of limitations and statute of repose calculated
[ ] CPSA § 15(b) reporting evaluation performed and documented

Pleadings and early motions

[ ] Every affirmative defense pleaded: preemption, misuse, alteration, comparative fault, assumption of risk, state of the art, compliance, repose, innocent seller, component-parts, successor limits
[ ] Chain-of-distribution and wrong-defendant challenges raised
[ ] Removal/forum analysis completed

Evidence build

[ ] Joint inspection protocol executed before any disassembly
[ ] Design history file, FMEA, testing, and standards compliance assembled
[ ] Complaint and warranty history audited and scoped
[ ] Privilege log built
[ ] 30(b)(6) designees identified and trained

Theory-by-theory attack

[ ] Manufacturing defect: unit tested against spec
[ ] Design defect: forum's test confirmed; RAD attacked on feasibility, cost, and offsetting hazards
[ ] Warnings: adequacy assessed; causation deposition taken; learned-intermediary or sophisticated-user defense invoked where available
[ ] State-of-the-art and compliance evidence marshalled
[ ] Plaintiff-conduct defenses developed and allocated

Dispositive and exposure

[ ] Daubert motions filed against design, causation, and warnings experts
[ ] Partial summary judgment sought theory by theory
[ ] Other-similar-incidents scope limited by order
[ ] Post-sale duty and recall exposure assessed
[ ] Indemnity and contribution claims tendered upstream
[ ] Punitive bifurcation sought; corporate-conduct narrative built

Run the whole sequence against the pretrial architecture in The Civil Pretrial Toolkit: From Pleadings to Summary Judgment and the subject-matter map in The Tort and Products Liability Toolkit.

Primary Authority

Restatements

Restatement (Third) of Torts: Products Liability §§ 1–4 (categories of defect; reasonable alternative design), § 10 (post-sale duty to warn), § 11 (duty to recall)
Restatement (Second) of Torts § 402A and cmts. j and k

Federal rules

Fed. R. Civ. P. 26(a)(2) (expert disclosure); Fed. R. Civ. P. 30(b)(6) (corporate designee); Fed. R. Civ. P. 37(e) (loss of ESI); Fed. R. Civ. P. 56 (summary judgment)
Fed. R. Evid. 401, 403 (relevance and prejudice — other similar incidents); Fed. R. Evid. 407 (subsequent remedial measures); Fed. R. Evid. 702 (expert testimony, as amended effective 1 December 2023)

Expert admissibility

Preemption and the government contractor defense

Regulatory

Consumer Product Safety Act, 15 U.S.C. § 2064(b) (substantial product hazard reporting); 15 U.S.C. § 2084 (reporting of certain civil actions); 16 C.F.R. Part 1115 (CPSC reporting interpretation)
General Aviation Revitalization Act of 1994 (18-year statute of repose for general aviation aircraft)

Punitive damages

BMW of North America, Inc. v. Gore, 517 U.S. 559 (1996); State Farm Mutual Automobile Insurance Co. v. Campbell, 538 U.S. 408 (2003); Philip Morris USA v. Williams, 549 U.S. 346 (2007)

State-law design tests (representative — confirm your forum)

Greenman v. Yuba Power Products, Inc., 59 Cal. 2d 57 (1963) (origin of strict products liability)
Barker v. Lull Engineering Co., 20 Cal. 3d 413 (1978); Soule v. General Motors Corp., 8 Cal. 4th 548 (1994)
Potter v. Chicago Pneumatic Tool Co. (Conn. 1997) (declining to require a reasonable alternative design in every design case)

Accuracy and currency notes.

1. This is state law. Products liability, and tort law generally, is overwhelmingly state law. The governing design-defect test (consumer expectations, risk-utility, hybrid, or a required reasonable alternative design), the availability of state-of-the-art, sealed-container, and innocent-seller defenses, the effect of substantial alteration, the comparative-fault regime, the existence and shape of a heeding presumption, statutes of repose, and punitive-damages caps all vary materially by jurisdiction. Insurance coverage — including what counts as timely notice and whether late notice requires a showing of prejudice — is likewise state law and varies. Nothing above states a nationwide rule except where it cites federal statutes, federal rules, or United States Supreme Court decisions. Confirm your forum's law before relying on any defense described here.

2. Chevron is no longer good law. Chevron U.S.A. Inc. v. Natural Resources Defense Council, Inc. was overruled by Loper Bright Enterprises v. Raimondo, 603 U.S. 369 (2024). Courts now decide questions of statutory interpretation de novo, with Skidmore-style respect for an agency's view to the extent it is persuasive. Any preemption or regulatory-compliance argument in a products case must be framed accordingly; a brief that asks a court to defer to an agency under Chevron is asking for relief that no longer exists.

3. Regulatory figures move. CPSC reporting guidance and timelines, civil penalty maxima (inflation-adjusted annually), consensus-standard revisions (ANSI, UL, ISO), and mandatory federal product standards change on their own schedules. Every regulatory threshold, deadline, and dollar figure in this checklist is illustrative and must be verified against the current statute, regulation, and agency guidance before you rely on it.

Secondary Sources

American Law Institute, Restatement (Third) of Torts: Products Liability (1998) — the modern organizing framework
Dobbs, Hayden & Bublick, The Law of Torts — the standard treatise on tort doctrine and defenses
Wright & Miller, Federal Practice and Procedure — Rule 26, Rule 37(e), Rule 56, and expert practice
Weinstein's Federal Evidence — Rules 403, 407, and 702
Uniform Comparative Fault Act (for the allocation framework adopted, with variations, in many states)

Related UpLaw Resources

Related Guides & Articles

Related Checklists

Related Toolkits

This checklist is general legal information from UpLaw (uplaw.us), not legal advice, and does not create an attorney-client relationship. Laws vary by state and change over time; verify current requirements and consult qualified counsel before acting.