Product liability insurance responds when a product you put into the market injures somebody or damages their property, and it covers both the settlement or judgment and the cost of defending the claim. For an importer, the critical fact is not the premium but the position: in most jurisdictions the entity that places the product on the market is treated as a producer in its own right, which means a customer with a claim will sue the brand or importer first and the factory not at all. Insurance is how that exposure is funded rather than how it is avoided.
This guide covers why the importer is usually the first defendant even when the defect was upstream, the three theories under which a claim is brought and what each requires the claimant to prove, what a claim realistically costs to resolve and why defence cost is the dominant number, why a policy limit is not the same as the capacity to pay, the six fields on a certificate that actually matter, what an additional insured endorsement transfers and what it leaves with you, why territory and jurisdiction exclusions defeat most supplier certificates, how retroactive dates create gaps when a supplier changes insurer, why recall costs are usually outside the policy, what to require in the supply contract, why a small brand needs its own cover as well as the supplier’s, which evidence decides how a claim ends, and how to keep a risk transfer file that works on the day it is needed. QUANZHOU JUNYUAN BAGS: custom waterproof bag production since 2014, 4,950 m² SGS-verified facility, MOQ 500 pieces per style, sampling in 6–10 working days and bulk in 35–50 days, FOB Xiamen.



Why the importer is usually the first defendant
A brand that buys a product liability insurance policy late usually does so after discovering an uncomfortable fact about the chain it sits in. A waterproof bag importer is not a passive intermediary in the eyes of a claimant’s lawyer. Under European product liability rules the importer into the union is treated as a producer, with the same liability as the manufacturer. Under United States law, every entity in the chain of distribution can be liable for a defective product, and the practical consequence is that the claimant sues whoever is easiest to serve and most able to pay.
That produces three specific consequences. The factory that made the bag is usually beyond the reach of the claimant’s jurisdiction, difficult to serve, and commercially inconvenient to pursue. The brand is local, has a website, has a marketplace storefront, has insurance, and has a reputation it would rather not litigate in public. And the marketplace or retailer in the middle will typically tender the claim to the brand under the indemnity in its own supplier agreement, which means the brand is defending a claim it did not cause and cannot control.
The second uncomfortable fact is that contractual indemnity from the factory does not solve this. An indemnity is a promise to pay, and its value depends entirely on the promisor’s ability and willingness to honour it, in a jurisdiction where enforcement is realistically available. A well-drafted indemnity from a supplier with no insurance and no assets in your market is worth approximately nothing, and it will not stop the claim being defended by you, in your market, at your cost, on your timetable.
This is why insurance and indemnity are complements rather than alternatives. The indemnity shifts the ultimate economic burden back up the chain. The insurance funds the defence while that shift is argued about, and it remains standing if the shift fails. A programme with one and not the other has a gap that only becomes visible when a letter arrives. The contractual side is set out in our guide to contract terms buyers should insist on.
The three theories: design defect, manufacturing defect, failure to warn
Almost every product claim is pleaded under one of three theories, and the theory determines what the claimant must prove and what evidence defends the case. Understanding which one applies to a given failure changes what a brand should keep on file, which is why this classification is practical rather than academic.
| Theory | What the claimant must show | Typical waterproof bag example | Evidence that defends it |
|---|---|---|---|
| Design defect | That the design itself is unreasonably dangerous, so that every unit is defective | A roll-top closure that opens under a load the product is marketed to carry, or a strap geometry that fails at a rated load | Design calculations, load test records, the reasoning behind the chosen rating, and evidence the design was validated against an actual duty |
| Manufacturing defect | That a particular unit or batch departed from the specification | One weld line with insufficient dwell, a batch of buckles from an unapproved source, a mis-set machine for one shift | Batch traceability, process control records, inspection results for that batch, and the ability to identify and segregate the affected units |
| Failure to warn | That a foreseeable risk was not adequately communicated | A dry bag marketed with water-sports imagery and no statement that it is not a flotation device, or no load limit on a strap | The instruction sheet as shipped, the warning on the product, and evidence the warning was tested for comprehension rather than drafted by a lawyer |
The manufacturing defect theory is the one importers find most frightening and the one that is actually most defensible, provided traceability exists. If you can identify which batch, which shift and which component lot a returned unit came from, and show that the process was in control for that batch, the claim narrows to a handful of units rather than to the whole line. If you cannot, the claimant will argue the defect is systemic — and without traceability there is no evidence to contradict that.
The failure to warn theory is the cheapest to defend against and the most commonly lost, because warnings are almost always written too late and too narrowly. A warning works when it addresses a foreseeable misuse, is visible at the point of use, and is in the language of the market. A sentence buried in a multilingual leaflet at the bottom of a polybag does not meet that standard. The specific misuses worth warning about in this category are use as a flotation or safety device, load limits on straps and handles, exposure to heat and flame, and the limits of submersion depth and duration.
A fourth cause of action sits alongside these three in most markets — breach of warranty — and it is worth knowing because it does not require anyone to have been injured. An express warranty is created by what you say about the product: a submersion rating, a load rating, a durability claim. If the product does not perform as stated, the claim exists without injury, and it is usually brought as a consumer complaint or a chargeback rather than as a lawsuit. That is the overlap with returns, and the numbers are discussed in our analysis of warranty and return rates.
What a claim actually costs, and why defence dominates
Buyers imagine a product liability claim as a large settlement. In practice the settlement is often the smaller number, and the cost that hurts is the cost of getting there. Two structural features drive this: liability defence is expensive per hour and slow, and most claims resolve by settlement rather than by judgment, which means the dominant variable is how long the other side believes you can afford to keep fighting.
- A modest contested claim with expert evidence routinely consumes a five-figure sum in defence costs before any settlement is discussed.
- A claim involving injury, fire or water damage to property can consume a six-figure sum and, in the worst cases, exceed a standard limit entirely.
- Marketplace and retailer claims are cheaper individually and more frequent, and they arrive as chargebacks rather than as lawsuits.
- The cost of internal time — retrieving records, reconstructing what happened, briefing lawyers — is real and rarely budgeted.
The asymmetry matters more than the absolute numbers. A small brand facing a claim it believes is meritless still has to respond, and responding costs money that a small brand may not have. That is the mechanism by which weak claims produce settlements: not because the claim is good, but because the defence is unaffordable. Insurance changes that calculation immediately, because it converts an unaffordable defence into a funded one, and funded defences settle for less.
There is a second-order cost that insurance does not cover and that buyers consistently underestimate: the commercial consequence. A retailer that experiences a safety issue may delist pending investigation regardless of the merits. A marketplace may suspend a listing. A distributor may withhold payment. Those losses are economic and generally outside a liability policy, which is why the preventive work — testing, inspection, documentation — is worth doing independently of the insurance question.
The United States recall and incident data published by the US Consumer Product Safety Commission gives a sense of how ordinary these events are across consumer categories, and the European product liability framework, recast so that member states apply it from late 2026, is available through EUR-Lex. Neither is comfortable reading for an importer with no cover.
Policy limits are not the same as capacity to pay
This is the single most important technical point in the whole subject, and it is the one buyers get wrong when they read a certificate. A limit is the maximum the insurer will pay. Capacity to pay is a function of whether that limit is available for your claim, in your jurisdiction, on a policy that is in force, with an insurer that can meet it. A large limit can sit behind any one of five gaps and be worth far less than it appears.
- Per-occurrence versus aggregate: a policy with a two million limit and a two million annual aggregate is exhausted by one large claim, leaving nothing for the rest of the year.
- Defence costs inside the limit: in many policies the cost of defending erodes the limit, so a long defence quietly consumes the money available to settle.
- Sub-limits and exclusions: a headline limit can be subject to lower sub-limits for particular heads of loss, and to exclusions that remove the exposure you actually have.
- Territory and jurisdiction: a limit that does not respond to suits brought in your market is not a limit for your purposes, however large the number.
- Insurer quality: a limit is a promise by an insurer, and the value of the promise depends on the promisor’s own standing and reinsurance.
The defence-costs point deserves a concrete illustration because it is the one that produces genuinely unpleasant surprises. Suppose a policy with a one million limit where defence costs are inside the limit. A claim runs for two years, defence costs reach several hundred thousand, and the amount left to settle with is the remainder rather than the headline figure. The same policy written with defence costs payable in addition to the limit behaves completely differently, and the difference is one clause that most people never read.
The practical consequence is that the limit should never be assessed alone. Three questions accompany it: is there a separate aggregate, and how large; are defence costs inside or outside the limit; and does the policy respond in the jurisdictions you sell into. A smaller limit that answers all three well is better cover than a larger one that fails any of them, and this is the comparison worth making rather than the one printed in bold on the certificate.
There is also a deductible or retention to consider, and it interacts with frequency. A high retention reduces premium but means the insured funds the first layer of every claim, which for a business with frequent small claims is the wrong trade. A business with frequent small property-damage or marketplace claims and rare large ones is usually better served by a lower retention and a higher limit than the reverse.
The six fields on a certificate that actually matter
A certificate of insurance is a summary, not a policy, and it exists to be read quickly. Six fields carry nearly all the information that matters, and reading them takes a couple of minutes. Everything else on the page is administrative.
| Field | What a good answer looks like | What a problem looks like |
|---|---|---|
| Named insured | The legal entity supplying your product, matching the name on your purchase order and invoices | A trading company, a parent, or a name that does not reconcile with any document you hold |
| Limits, split out | A per-occurrence limit and a separate, larger annual aggregate, both stated | A single number with no aggregate, or an aggregate equal to the per-occurrence limit |
| Effective and expiry dates | Current, covering the period of your shipments and re-checked annually | Expired, or a certificate issued once and reused for years |
| Territory and jurisdiction | Expressly includes the countries you sell into, including the United States and Canada if you sell there | Worldwide excluding the USA and Canada, which is the standard exclusion in many non-US programmes |
| Products and completed operations aggregate | Shown separately and at a meaningful level | Absent, or so low that it is exhausted by a single claim |
| Cancellation notice and additional insured | A stated notice period to the certificate holder, and your entity named as additional insured | No notice provision, or an additional insured box that is blank on the copy you were sent |
The territory row deserves emphasis, because it is the one that silently defeats the majority of supplier certificates held by importers selling into North America. Many programmes written outside the United States exclude suits brought in the United States and Canada, whether by a territory clause or by a jurisdiction clause. The certificate still looks impressive, still shows a large limit, and will not respond to the claim that is actually most likely to arrive. Reading the territory clause before accepting a certificate takes seconds.
The named insured row catches a different and equally common problem. Certificates are often issued to a group parent or to a trading company, while the purchase order and the invoice name a different legal entity. If the entity that made the product is not the entity insured, an additional insured endorsement attached to the wrong policy is of no use. Reconciling the three names — certificate, contract, invoice — is a five-minute check that prevents a structural failure in the whole arrangement.
One more administrative discipline pays for itself: diarise the expiry. Certificates are usually issued annually, and the most common state of affairs in a mature supply relationship is a certificate that expired eighteen months ago and nobody noticed. Requiring a fresh certificate with each annual review, and filing it against the season, costs nothing and removes the entire class of problem. Supplier document discipline generally is covered in our supplier audit checklist.
What an additional insured endorsement transfers, and what it leaves with you
Being named as an additional insured on a supplier’s policy is the most commonly requested and least well understood protection in a supply contract. It does something specific and valuable: it gives you rights under their policy for liability arising out of their product. It does not make you immune, it does not mean their insurer will defend you first, and it does not transfer the whole exposure.
- It is usually limited to liability arising out of the named insured’s products or work. Your own independent negligence is generally not covered.
- It can be primary or excess. If it is excess, it responds only after your own policy is exhausted, which changes its practical value substantially.
- It is subject to all the terms of the underlying policy, including the territory and jurisdiction exclusions discussed above.
- It does not prevent you being sued. It provides funding for a claim that is still brought against you and still appears on your record.
- It can be defeated by late notice. Most policies require prompt notification, and a claim discovered months late can prejudice cover.
The primary-versus-excess distinction is the one worth insisting on in contract language. An endorsement that is "primary and non-contributory" means the supplier’s policy responds before yours and without calling on yours for contribution, which is the arrangement that actually protects your own loss record. An endorsement that is silent on the point may be construed as excess, which means your policy pays first and your claims history takes the hit even though the defect was upstream.
Notice is the second practical gap and it is entirely within the buyer’s control. Liability policies require the insured to notify claims or circumstances promptly, and an additional insured has its own obligations in many wordings. A brand that receives a complaint, sits on it for six months while deciding what to do, and then notifies, has created an argument for the insurer that would not otherwise exist. The control is procedural: any written complaint alleging injury or property damage goes to the broker immediately, whether or not it looks serious.
There is one thing an additional insured endorsement does that is genuinely valuable beyond the money: it gives you access to the supplier’s insurer, which means the supplier has a commercial incentive to cooperate with the defence. An uninsured supplier faced with an indemnity claim has every incentive to deny involvement, and no incentive to share records. That cooperation is often worth more than the limit, because the defence depends on documents only the supplier holds.
Territory, jurisdiction and the exported product problem
Insurance is written for a market, and a policy that looks global frequently is not. The clause that decides the question is the territory clause, sometimes reinforced by a jurisdiction clause, and the two together determine whether a suit brought in your country is covered at all. For importers this is the highest-yield field on the certificate and the one most often left unread.
- Territory defines where a claim may be brought. A policy limited to the supplier’s domestic territory will not respond to a suit filed where you sell.
- Jurisdiction can exclude the United States and Canada specifically, even where the territory clause is broad, because of the cost and unpredictability of those courts.
- Some programmes cover worldwide territory but exclude any product knowingly exported to a named market, which is a different and subtler exclusion.
- Local admitted versus non-admitted cover matters in some markets, where a policy from an unauthorised insurer may not be recognised without a local fronting arrangement.
The knowingly-exported variant is worth naming because it catches people who did read the territory clause. A supplier may hold broad worldwide cover while a separate exclusion removes products the insured knew were destined for a named market. If the commercial arrangement involves that market, the exclusion applies, and the only way to find it is to ask for the policy wording rather than the certificate. Certificates rarely reproduce exclusions in full.
The remedy is straightforward and should be written into the contract: the supplier’s product liability cover must expressly include the territories named in the schedule of markets, and the supplier must provide the policy wording or a broker’s confirmation to that effect, not merely a certificate. A broker’s letter confirming that the programme responds to suits brought in the named jurisdictions is a reasonable and commonly provided document, and it costs the supplier nothing to obtain.
Where a supplier genuinely cannot obtain cover for your market, the honest position is that the risk remains with you. That is not a reason to abandon the relationship; it is a reason to price your own policy accordingly and to treat the supplier indemnity as worth less than it appears. Importers who understand this tend to buy more of their own cover rather than spending more time negotiating someone else’s. The destination-market picture is summarised in our overview of certification requirements across global markets.
Retroactive dates and the gap after a supplier changes insurer
Most product liability cover in a commercial package is written on an occurrence basis, which responds to an event during the policy period regardless of when the claim arrives. Some programmes, however, are written on a claims-made basis, which responds only if the claim is made during the period and relates to an event after a stated retroactive date. Where claims-made wording applies, the retroactive date becomes one of the most consequential fields on the document.
- A retroactive date set at the start of the current policy means events before that date are not covered, even if the claim arrives today.
- A supplier that changed insurer last year may now hold a policy whose retroactive date post-dates the shipments you are most worried about.
- A gap between policies, even a short one, can leave a period permanently uncovered under claims-made wording.
- Run-off or extended reporting cover exists to fill these gaps, but it has to be bought deliberately and usually within a short window.
The practical risk is specific and worth stating plainly. Goods shipped in one year generate claims in a later year, because a bag is used for seasons before anything goes wrong. A supplier that switched insurer in the interval, onto a claims-made policy with a fresh retroactive date, holds cover that does not respond to the earlier shipments. The certificate looks current, the limit looks adequate, and the period of exposure is excluded. Nothing on the front page discloses this.
The control is two questions asked at the same time as the territory question: is the cover occurrence-based or claims-made, and if claims-made, what is the retroactive date. A supplier with continuous occurrence cover answers the first question and the second becomes irrelevant. A supplier with claims-made cover should be asked to confirm continuous retroactive cover dating back to the start of the relationship, which a broker can usually arrange or confirm.
This interacts with a second insurance discipline that importers control directly: their own continuity. A brand that switches insurer every year on price, without checking that the new policy reaches back, can create exactly the same gap in its own cover. Buying your own policy is not a one-off transaction; it is a continuous chain, and the chain matters more than any single year’s premium. The goods-in-transit side of the same continuity question is handled differently, and is set out in our guide to cargo insurance claims.
Recall costs are usually outside the policy
One of the most expensive surprises in this subject is the discovery that standard product liability cover does not pay for a recall. Liability insurance responds to third-party injury and property damage. The cost of withdrawing a product that has not injured anyone — notification, returns handling, replacement stock, freight, marketplace delisting consequences — is a first-party economic loss, and it is commonly excluded or available only under a separate recall endorsement.
- Third-party cover responds to injury and damage to other people’s property caused by the product.
- First-party recall cost is the insured’s own loss and generally needs its own insuring clause.
- Recall endorsements vary widely in trigger: some require a government order, others respond to a voluntary withdrawal on professional advice.
- Some markets treat the mere announcement of a safety issue as a covered event under specific regulatory regimes, which is a narrower and different trigger again.
The distinction between a voluntary withdrawal and a mandated recall is where programmes get caught. A brand that discovers a defect, acts responsibly and withdraws the product before anyone is hurt has done the right thing and may find the cost entirely uninsured, because nothing has triggered third-party liability. A brand that waits until somebody is injured has a covered claim and a much worse outcome. That is a perverse incentive embedded in the standard wording, and it is worth knowing before the decision arises.
For most small and mid-sized importers, the proportionate response is not to buy an expensive recall product but to reduce the probability of needing one, and to know the number. Reducing probability means testing and inspection; knowing the number means calculating what a withdrawal of one season’s shipments would cost, including freight, replacement and marketplace consequences. That number is usually larger than expected, and it argues for spending more on prevention. The quality controls that reduce it are described in our complete guide to quality control and inspection.
Where recall cover is purchased, the wording to look for is the trigger and the definition of covered costs. A policy that responds only to a government-mandated recall will not help with the voluntary withdrawal that is the more likely scenario, and a policy that covers notification but not replacement stock only addresses part of the loss. Those are the two lines to read, and they are usually in the extensions section rather than the main wording.
What to require from a supplier, in contract language
A requirement stated informally produces a certificate, which is a summary. A requirement stated in the contract produces a programme, which is a set of obligations. The following clauses are standard, unremarkable to a competent supplier, and worth insisting on before the first order rather than after the first incident.
- Minimum limits stated as a per-occurrence figure and a separate, larger annual aggregate, with a figure appropriate to the markets you sell into.
- The buyer named as additional insured, on a primary and non-contributory basis, for liability arising out of the supplier’s products.
- Territory and jurisdiction expressly including every market named in the schedule, with written confirmation from the broker rather than a certificate alone.
- Waiver of subrogation in favour of the buyer, so the supplier’s insurer does not pursue the buyer after paying a claim.
- A stated notice period to the certificate holder before cancellation or material change, so cover cannot lapse silently.
- A fresh certificate before the first shipment and annually thereafter, filed against the season, with expiry dates recorded.
- An indemnity in the buyer’s favour, sitting alongside the insurance rather than instead of it.
The waiver of subrogation is the clause most often omitted and it has a specific job. Without it, the supplier’s insurer pays a claim and then exercises its right to recover from whoever else was responsible, which can include the buyer. With it, that right is given up, and the arrangement genuinely transfers rather than circulates the loss. Most suppliers agree to it without difficulty because their own insurer is usually content to include it, and it is a single line.
The limits clause is worth a note on calibration rather than a number. Requirements vary by market and by retailer, and national retailers in several markets stipulate a minimum that is commonly in the low single-digit millions. The right approach is to ask your largest customer what they require of you and to pass a comparable requirement down the chain, rather than inventing a figure. That way the requirement is calibrated to the exposure you actually carry rather than to a number copied from a template.
One clause deserves to be added even when the supplier resists it, because it costs nothing: a duty to notify the buyer of any complaint, claim or safety issue concerning the product, promptly and in writing. Late knowledge is what turns a manageable issue into a failure-to-warn claim, and a contractual notification duty creates a record that the buyer asked to be told. Where a defect later emerges, that record is the difference between a party that responded and a party that knew.
Why the small brand needs its own policy too
The error that costs small importers the most is the belief that the supplier’s insurance covers them. It does not, for the reasons already set out: the territory may exclude your market, the endorsement may be excess, the named insured may be the wrong entity, and none of it prevents you being the defendant. A brand selling one container a year still needs its own cover, and the premium is usually far lower than the number people imagine.
- Your own policy responds in your market, to your claims, on your timetable, without arguing about whose policy applies.
- Marketplace agreements frequently require commercial general liability cover at a stated level once sales pass a threshold, and may require proof on request.
- Retailers and distributors often require the brand to carry cover at a stated limit as a condition of listing.
- A funded defence changes settlement dynamics immediately, which is worth more than the limit itself for a business with limited cash.
The premium question is worth addressing directly because it is the usual objection. For a small importer with modest revenue, a straightforward products liability policy is a manageable annual cost, and the variables that move it are the limit, the retention, the territories and the product category. Bags are a relatively benign category compared with many, which helps. The cost is best compared not with zero but with the defence cost of a single modest claim, against which it is small.
There is a specific trigger new brands should know about. Marketplace seller agreements commonly require commercial general liability insurance at around one million once a sales threshold is passed, with proof provided on request. That requirement arrives by email, usually without warning, and brands without cover have to arrange it under time pressure. Anticipating it and having the policy in place before the threshold is crossed is cheaper and less stressful, and it means the cover is already running when it is needed.
Finally, there is a strategic reason that has nothing to do with claims. Holding your own policy means you have an insurer, a broker and a claims procedure before anything goes wrong. When an incident occurs, the first seventy-two hours determine how it is handled, and having somewhere to call is worth a great deal. Brands that buy cover after an incident discover that it does not respond to an event that has already happened, which is the one thing insurance can never do. The wider commercial framing is covered in our guide to building a waterproof bag brand.
The evidence that decides how a claim ends
Product claims are won and lost on documents, and almost all of the documents are ones a competent programme already keeps for other reasons. The difference between a defensible file and an indefensible one is rarely the size of the company; it is whether the records connect to a batch. Five categories carry the weight.
| Evidence | What it establishes | Why it decides outcomes |
|---|---|---|
| The specification and tech pack | What the product was supposed to be | Without it, there is no baseline, and a manufacturing defect cannot be distinguished from a design defect |
| Batch traceability | Which units came from which run, shift and component lot | Narrows a claim to a defined population instead of the whole line, and enables a proportionate remedy |
| Test and inspection records, dated | That the product was checked and how | Demonstrates reasonable care, and shows whether the issue was detectable and whether it was detected |
| The complaint register | What was known, and when | The most dangerous document in any claim, because prior complaints without action convert a defect into a known defect |
| Instructions and warnings as shipped | What the user was actually told | Decides failure-to-warn claims, which are the cheapest to prevent and the easiest to lose |
The complaint register deserves the most attention, because it is the document that turns a bad situation into an indefensible one. A handful of earlier complaints about the same failure, sitting unanswered in a marketplace inbox, is evidence that the defect was known and nothing was done. That converts a manufacturing defect argument into something considerably worse and, in some jurisdictions, opens the door to a very different scale of damages. The control is unglamorous: log every complaint, with a date and a disposition, and act on patterns.
The second document that decides cases is the internal email. Claimants obtain disclosure, and the most damaging material is rarely the test report; it is a message in which somebody identified a problem and the business decided to continue shipping. The cultural control is simple and worth stating to anyone in the business: write as though the document will be read in court, because it might be, and never record a decision to ship something known to be defective.
Positive evidence is worth building deliberately too. A dated test report on the specific construction, an inspection record for the specific batch, a specification signed before production, and photographs of the batch being packed together cost very little and convert a defence from assertion into proof. Programmes that already do this for quality reasons discover they have been building a liability file all along. The testing that produces it is described in our review of third-party testing laboratories.
Keeping a risk transfer file that works on the day
A risk transfer arrangement is only as good as the file that evidences it, and the file is only useful if it can be produced in an afternoon. Everything in this guide reduces to a folder with a small number of sections, maintained once a year, that anybody in the business can find. Building it takes a day and maintaining it takes an hour a year.
- Contracts: the supply agreement with the indemnity and insurance clauses, signed and current.
- Certificates: a current certificate from every supplier, with expiry dates recorded and diarised.
- Confirmations: broker or insurer confirmation of territory, jurisdiction and retroactive cover where relevant.
- Your own policy: the policy wording, not just the schedule, with the limit, aggregate, retention and defence-cost position noted.
- Technical file: specification, test reports and inspection records, filed by season and by batch.
- Complaints: a register with dates and dispositions, and the corrective actions taken.
- Contacts: the broker, the insurer’s claims line, and the named person at each supplier who handles a claim.
The contacts line is the one that sounds trivial and is not. When an incident occurs, the first hours go into finding out who to call, and that delay is where mistakes happen. A single page with the broker’s name, the claims notification address, and the escalation contact at each supplier turns a scramble into a procedure. It is also the page most likely to be out of date, so it should be refreshed at the same annual review as the certificates.
One annual discipline ties the whole thing together and takes under an hour. Once a year, before the peak season, check that every supplier certificate is in date, that the named insured still matches the invoicing entity, that the territory still covers the markets you now sell into, and that your own policy is renewed on continuous terms. Four checks, one hour, and the difference between a risk transfer arrangement that exists and one that works.
If you want this applied to a specific programme, send the markets you sell into, your annual volume and the largest customer you supply, and the insurance requirement can be calibrated against those three before the next order is placed. You can see how a programme moves from first enquiry through sampling into bulk production; every style starts at 500 pieces minimum, with samples in 6–10 working days and bulk in 35–50 days, quoted FOB Xiamen.
Frequently Asked Questions
Q1. Why is the importer liable if the factory caused the defect?
Because most regimes treat the entity that places a product on the market as a producer in its own right, and because the importer is the party that is local, solvent and easy to sue. Contractual indemnity shifts the burden, but only if it is enforceable.
Q2. What are the three product defect theories?
Design defect, where the design itself is unreasonably dangerous; manufacturing defect, where a unit or batch departed from the specification; and failure to warn, where a foreseeable risk was not adequately communicated.
Q3. Which theory is easiest to defend?
Manufacturing defect, provided batch traceability exists. Being able to identify the affected run narrows the claim to a defined population rather than the whole product line.
Q4. Does a supplier indemnity replace insurance?
No. An indemnity is a promise to pay, and its value depends on the promisor being able and willing to pay in a jurisdiction where you can enforce. Insurance funds the defence while that is argued.
Q5. Why is a policy limit not the same as capacity to pay?
Because the limit may be eroded by defence costs, capped by a small aggregate, excluded by territory, or underwritten by an insurer that cannot meet it. A smaller limit that answers all four is better cover than a larger one that fails one.
Q6. Are defence costs inside or outside the limit?
It depends on the wording, and it matters enormously. Where defence costs erode the limit, a long defence quietly consumes the money available to settle.
Q7. What does an additional insured endorsement actually do?
It gives you rights under the supplier’s policy for liability arising out of their product. It does not stop you being sued, and it is usually limited to their liability rather than your own negligence.
Q8. Why does primary and non-contributory matter?
Because an endorsement that is silent may be treated as excess, so your own policy pays first and your claims history takes the hit even though the defect was upstream.
Q9. What is the biggest gap in supplier certificates?
Territory. Many programmes exclude suits brought in the United States and Canada, so a large limit does not respond to the claim most likely to arrive.
Q10. What is a retroactive date?
On a claims-made policy it is the date from which events are covered. A supplier that changed insurer may hold cover that excludes the earlier shipments that generated the claim.
Q11. Does product liability insurance pay for a recall?
Usually not. Standard cover responds to third-party injury and property damage. The cost of withdrawing a product is a first-party loss and generally needs a separate recall endorsement.
Q12. What is the perverse incentive in standard recall wording?
A voluntary withdrawal before anyone is hurt may be entirely uninsured, while waiting for an injury creates a covered claim. Knowing that in advance changes how the decision is made.
Q13. Does a small brand really need its own policy?
Yes. Marketplace and retailer agreements commonly require it, and a funded defence changes settlement dynamics immediately. The premium is small against the defence cost of one modest claim.
Q14. What limit should I require from a supplier?
Calibrate it to what your largest customer requires of you, rather than copying a template. Ask them, then pass a comparable requirement down the chain.
Q15. What document most damages a product liability defence?
A complaint register showing earlier complaints about the same failure with no action taken, because it converts a defect into a known defect.
Q16. What should be in a risk transfer file?
Signed contracts with insurance and indemnity clauses, current supplier certificates with expiry dates, territory confirmations, your own policy wording, the technical file by batch, a complaint register, and contacts.
Q17. How often should insurance documents be reviewed?
Annually, before peak season. Check that certificates are in date, that the named insured still matches the invoicing entity, that territory still covers your markets, and that your own cover is continuous.
People Also Ask
Who is liable for a defective imported product?
Usually the importer or brand, because most regimes treat the party placing the product on the market as a producer and that party is easiest to sue.
What are the three types of product defect?
Design defect, manufacturing defect and failure to warn. Each requires different evidence and each is defended differently.
Does a supplier indemnity replace insurance?
No. An indemnity is a promise to pay; insurance funds the defence and remains standing if the promise cannot be enforced.
Does product liability insurance cover a recall?
Usually not. Recall cost is a first-party loss and generally requires a separate endorsement with its own trigger.
Why is territory important on a supplier certificate?
Because many programmes exclude suits brought in the United States and Canada, so the limit does not respond where the claim arrives.
Do small importers need their own product liability policy?
Yes. Marketplaces and retailers commonly require it, and it funds the defence that decides how a claim settles.