Limitation of liability clauses: what to do in the first ten days comes down to three moves: read the clause against the specific claim that has arisen, preserve the record before anyone drafts a position, and identify who carries the burden of showing whether the cap applies. Delay beyond that window narrows the options still open.
Who this concerns
The question surfaces the same way almost every time. A claim lands, someone opens the contract to the limitation clause everyone signed off on months or years earlier, and the clause turns out to say less than the party invoking it assumed. This is rarely a drafting failure that only shows up later. It is a gap between what the clause was meant to do and what it actually says once tested against a real dispute rather than a hypothetical one.
Commercial counterparties on both sides of a Swedish contract face this. The supplier trying to cap exposure on a delivery failure, the buyer trying to push past a cap that looks designed to absorb exactly the loss that has occurred, the party in a warranty dispute checking whether a cap in a purchase agreement survives a claim framed as fraud rather than breach. Whoever moves first, with the clearer record, sets the terms of the argument that follows.
Ten days is not a statutory deadline. Nothing in Swedish contract law requires a response within that window. It is the practical horizon within which the record is still fresh, witnesses still remember the sequence of events, and the counterparty has not yet had time to build a narrative that forecloses a reading of the clause that favours the other side. Past ten days none of that disappears outright, but each day makes the eventual negotiation or dispute marginally harder to control.
What the law says
Under Swedish law as it currently stands, limitation of liability clauses agreed between commercial parties of broadly comparable bargaining strength are enforced largely as written. Freedom of contract is the starting point, not the exception, and Swedish courts do not rewrite a negotiated allocation of risk simply because one side now regrets the terms it agreed to.
That latitude is not unlimited. A cap invoked against conduct amounting to gross negligence or wilful default typically does not hold, regardless of how the clause is worded, because that conduct falls outside what the parties can be taken to have contemplated when they agreed to limit exposure. A cap can also be set aside, in whole or in part, where enforcing it as written would produce a result manifestly unreasonable given how the dispute has actually unfolded. This is an exception applied narrowly, not a general escape route from an inconvenient clause.
None of this changes what happens in the first ten days. The legal test that eventually applies to the clause is decided later, often much later, by a court or an arbitral tribunal. What gets decided in the first ten days is the evidentiary and procedural position the case will run on once that test is applied. A weak position on day ten is rarely repaired by a strong legal argument two hundred days on.
How it works in practice
Day one: read the clause against the claim that has actually arisen
Most limitation clauses are drafted to a generic loss profile, not the specific claim now on the table. Check whether the cap applies to the type of loss claimed at all, whether it is a single aggregate cap or a per-event cap, whether indirect and consequential loss are carved out separately from the cap, and whether the clause survives termination of the underlying agreement or falls away with it.
Days two and three: preserve the record before anyone drafts a position
Internal emails, delivery logs, inspection reports and prior correspondence about the same performance issue matter more at this stage than any legal argument. Once a dispute posture forms, people stop writing candidly. The window in which the file still reflects what actually happened, rather than what someone wishes it had said, is short.
Days four to six: map who carries the burden of proof
In most drafting, the party seeking to rely on the cap does not have to prove the cap applies; the party trying to break it has to prove the exception does. That allocation is not always obvious from the clause's wording and is worth confirming early, because it determines which side needs to build a factual case first.
Days seven and eight: check notice and time-bar provisions in the contract itself
Many limitation clauses are paired with a separate notice requirement: a claim must be raised within a defined period after the loss becomes apparent, failing which the right to claim at all, not just the amount, is affected. This is a contractual mechanism, not a statutory one, and it is set out in the agreement rather than in general law.
Day nine: test whether the clause survives contact with mandatory rules
Some categories of loss cannot be capped regardless of drafting, and some counterparties are protected by rules that override a negotiated allocation of risk. Before treating the cap as the ceiling on exposure, confirm the claim does not fall into one of those categories.
Day ten: decide whether to preserve, negotiate, or escalate
By day ten the practical choice is usually one of three: hold the position and prepare for a longer dispute, open a negotiation from a documented position rather than a defensive one, or escalate internally because the exposure exceeds what can be resolved commercially. The record built in the previous nine days determines which of these is realistic.
Where the counterparty sits outside Sweden
A foreign counterparty, foreign assets, or a foreign parent company change the calculation. A cap that would be respected between two Swedish commercial parties may be tested against a different governing law if the contract says so, or against a different forum's approach to gross negligence exceptions if enforcement ends up abroad. Where assets sit outside Sweden, the practical value of winning an argument about the cap depends on whether a resulting judgment or award can actually be enforced against those assets, which is a separate question from whether the clause holds as a matter of contract interpretation.
What to check before day ten runs out
- Whether the cap is aggregate, per-claim, or tiered by category of loss
- Whether indirect or consequential loss is defined in the contract or left to general interpretation
- Whether a separate notice period applies and, if so, when the clock on it started running
- Whether the clause survives termination or expires with the agreement
- Who, under the drafting, carries the burden of proving the exception applies
- Whether governing law and forum clauses point outside Sweden
Can settlement or mediation change how a limitation of liability clause is applied?
Yes, though not by rewriting the clause. In a mediated settlement the parties can agree to depart from the cap entirely, trading certainty for a number both sides can live with. What mediation does not do is resolve the underlying question of whether the cap would have held in litigation; that question is simply set aside. Anyone weighing settlement and mediation mechanics against a contested cap should read that position alongside the clause itself, not instead of it.
How does inlösen interact with a limitation of liability clause?
Inlösen, the compulsory buyout mechanism available once one shareholder holds enough of a company, sits outside the contractual limitation regime entirely. It compensates the minority at a value set by rules separate from whatever cap the parties negotiated in a supply or services agreement. A limitation clause has no bearing on an inlösen valuation, and treating the two as connected is a common misreading.
Does a force majeure clause override a limitation of liability clause?
Not automatically, and the two clauses answer different questions. A force majeure provision excuses non-performance triggered by an event outside a party's control. A limitation of liability clause caps what is owed once liability exists at all. Where a contract addresses both, the drafting on force majeure and changed circumstances is checked first. If it excuses performance, the limitation clause is never reached.
The numbers
Ten days is an operational horizon, not a legal one. Nothing in Swedish contract law sets that figure; it reflects how quickly a record degrades once a dispute posture forms. The general limitation period for contractual claims under Swedish law is measured in years, not months, but that longer horizon has almost no bearing on how the first ten days should be used, since the position built now is what the eventual claim, if there is one, will rest on.
Notice periods, where they exist, come from the contract itself rather than from statute, and they are typically expressed in days from the point the loss becomes, or should have become, apparent. What increases the eventual cost of a dispute over a cap is rarely the clause itself; it is the quality of the record built in the days immediately after the claim arose, and whether the counterparty's own file is better organised than yours by the time either side needs it.
An industry example illustrates the point without needing invented figures: contracts on long-lead infrastructure work, of the kind discussed in the piece on permitting an industrial energy project, often stack a limitation clause on top of milestone-based delay damages, and the interaction between the two is worked out from the drafting, not from a general rule.
Where it usually goes wrong
The most common failure is treating the cap as fixed when the clause actually distinguishes between categories of loss, some capped and some not. A close second is assuming the clause survives termination when the drafting says otherwise, or the reverse: assuming it lapses when in fact it was drafted to survive.
Gross negligence exceptions are invoked more often than they succeed. The bar for showing conduct falls outside ordinary negligence is higher than commercial instinct suggests, and building a case around that exception without a genuinely strong factual record wastes the ten-day window on an argument unlikely to hold.
Foreign law and forum clauses are underestimated. A cap negotiated on the assumption that Swedish courts would apply Swedish reasonableness standards can behave very differently once the contract's own forum clause routes the dispute elsewhere, or once enforcement has to happen against assets sitting in a jurisdiction with a different approach to limitation clauses altogether.
Finally, indemnities are frequently conflated with the limitation clause. A separate indemnity for a defined category of loss, such as breach of specific warranties, often sits outside the general cap by design, and missing that distinction leads parties to argue the wrong clause entirely.
What to do next
The work described above, reading the clause against the claim, preserving the record, and mapping the burden of proof, is work a commercial team can and should do internally in the first ten days. Where it stops being self-directed is the point at which the clause's wording is genuinely ambiguous against the facts, or where the counterparty sits in another jurisdiction and enforcement becomes part of the calculation. That is where an outside assessment of the position, rather than a general read of the clause, becomes worth commissioning.
For transactions where the limitation clause sits inside a wider purchase agreement, the mechanics of how caps and indemnities are structured are covered in more detail on the page dealing with asset deal transfer undertakings. For everything else within the contracts and transactions practice, a short assessment call is the fastest way to find out whether the position built in the first ten days holds up, and what it would take to test it. Requests go through the firm's contact page and receive a response within two hours on a working day.