Two general liability policies can carry the exact same coverage and still respond completely differently to the same claim — because they are triggered differently. One is written on an occurrence basis, the other on a claims-made basis, and for a machine shop or manufacturer the difference is not academic. A defective product can surface a claim a long time after it ships, and which trigger your policy uses decides which policy year answers — or whether any of them does. This post explains the two triggers in plain terms, why the long products tail makes the choice matter so much for a manufacturer, and what to watch so a late-arriving claim does not find a gap.
The short version: occurrence coverage triggers on when the harm occurs; claims-made coverage triggers on when the claim is reported. Both are ways of triggering the products-completed operations side of your general liability policy. Because a manufacturer’s products keep causing claims long after they leave the dock, the occurrence trigger usually fits the exposure better — but the right answer depends on your situation and the form your policy actually carries.
The two triggers, in plain terms
Almost every commercial general liability policy is published in one of two standard ISO coverage forms, and the form is the trigger.
An occurrence policy — typically the form known as CG 00 01 — responds to bodily injury or property damage that occurs during the policy period, no matter when the claim is finally made. If the harm happened while the policy was in force, that policy answers, even if the claim arrives years later, after the policy term has closed. The trigger is the event, not the paperwork.
A claims-made policy — the CG 00 02 version — responds based on when the claim is reported instead. The harm and the claim are treated together: the policy in force when the claim is made is the one that answers, subject to a retroactive date that sets how far back the coverage reaches. The trigger is the report.
Same general liability coverage. Same products-completed operations hazard underneath it. Two different answers to the question of which policy year picks up a claim. Forms, editions, and endorsements vary by carrier, so the practical move is to read which form your policy is built on rather than assume.
Why the products tail makes the trigger matter
For most businesses the occurrence-versus-claims-made distinction is a detail. For a manufacturer it is close to the center of the program, and the reason is the tail.
Products-completed operations is a long-tail exposure. A part you machined or a product you made keeps existing — installed, used, resold, relied on — long after it leaves your control, and a defect in it can turn into a third-party injury or property-damage claim months or years after the sale. That delay between the product shipping and the claim arriving is exactly the gap the trigger governs. When the lag is short, it barely matters which trigger you have. When the lag can run for years, the trigger decides everything about which policy answers.
What occurrence buys a manufacturer
The appeal of occurrence-based coverage for a shop or plant is continuity you do not have to manage. Because the trigger is the moment of harm, the policy that was in force when the product failed is the one that answers — even if the claim shows up long after that policy year ended, and even if you have changed carriers since. Each policy year effectively locks in coverage for the products that were out in the field during it. For a manufacturer whose output keeps causing potential claims for years, that lock is worth a great deal, because it means a late claim looks back to the right year on its own.
That is why the standard occurrence form, CG 00 01, is so common for this class. It matches the shape of the exposure: harm now, claim later, coverage anchored to the harm. It is also why the products-completed operations side of the policy and its separate aggregate are read together — the trigger decides which year’s aggregate a claim draws against, and the aggregate decides how much is left in that year to pay it.
What claims-made demands in return
Claims-made coverage is not worse coverage, but it asks more of you to keep it intact, and for a long-tail class that maintenance is the catch.
Two things have to be watched. The first is the retroactive date — the cutoff for how far back the policy will reach. A claim reported during the term is generally only covered if the harm occurred on or after that date, so if the retroactive date drifts forward (often when coverage lapses or is rewritten), older production can fall outside the policy even while you are still insured today. The second is continuity. Because the policy answers based on reporting, a gap between policies — or a switch to a new form without a bridge — can leave harm that occurred in the covered years uncovered when the claim finally arrives. Tail coverage, an extended reporting period, exists precisely to bridge that gap, and it is a conversation to have before you change forms or carriers, not after.
Real-World Scenario: A shop ships a run of components, and three years later one fails in the field and injures an operator — the claim lands now. On an occurrence policy, the year the harm occurred answers, full stop, even though that policy term closed long ago and even if the shop changed carriers twice since. On a claims-made policy, the answer turns on the policy in force today, its retroactive date, and whether coverage stayed continuous across those three years. Same failure, same product — and a very different path to a covered claim depending on the trigger.
How to tell which trigger you carry
The fastest way to know how your policy will respond is to read the coverage form it is built on. Your declarations page or the policy form schedule names the general liability coverage form by number, and the occurrence form and the claims-made form carry different identifiers — the occurrence version typically the CG 00 01, the claims-made version the CG 00 02. If the policy is claims-made, the declarations will also show a retroactive date, which an occurrence policy does not carry in the same way; seeing that date on the page is itself a strong signal of the trigger. Endorsements and editions can modify the details and vary by carrier, so the form number is the starting point rather than the whole answer.
If you are not sure what you are looking at, that is exactly the read we do for a shop or plant before binding — because finding out which trigger you carry during a claim is the most expensive time to learn it. We check the form, the retroactive date if there is one, and whether continuity has held across renewals and carrier changes, so the products you shipped in prior years are accounted for rather than assumed.
Why it matters for your operation
For a machine shop or manufacturer, the occurrence-versus-claims-made choice is really a question about the tail: how long after a product ships can a claim still arrive, and do you want coverage anchored to the harm or to the report. Most of the time the long products tail makes occurrence-based coverage the better fit, but the only way to know what you have is to read the form your policy is actually built on — and to manage the retroactive date and continuity carefully if it is claims-made. We check the trigger against the products you ship, alongside the products liability that lives inside general liability, the separate products aggregate behind it, and how the three products coverages fit together, so a late claim does not find a gap. When you are ready, start a quote and tell us what you make or machine, or read the full general liability and products liability page. Forms and editions vary by carrier, so the right move is always to confirm which trigger your policy actually carries rather than assume.