Liability Caps in Contracts, Explained (2026)

A liability cap is the maximum amount one party can be required to pay the other if things go wrong under a contract. It sits inside the limitation of liability clause and it is often the single most negotiated number in a commercial deal, because it decides how much is really at stake. This explainer sets out what a cap is, how it is usually set, the carve-outs that sit outside it, and what the law allows. It is general information, not legal advice; enforceability varies by country. ## What a liability cap is When a contract is breached and causes loss, the law would normally let the injured party recover its provable, foreseeable damages with no fixed ceiling. A liability cap overrides that by agreement: the parties decide in advance the most that can be claimed. For example, "the total liability of each party under this agreement shall not exceed the fees paid in the 12 months before the claim." If the loss is larger than the cap, the injured party simply cannot recover the excess from its counterparty. ## How caps are usually set There is no single market rule, but common approaches are: - Fees paid: the total fees paid under the contract, often measured over the 12 months before the claim. This is the most common approach in services and SaaS.

• A multiple of fees: 2x or 3x annual fees, used where the customer wants more cover than a single year.
• A fixed sum: a stated figure, sometimes tied to the value of the deal or to the limit of the supplier's insurance.
• Contract value: the total price of a one-off project. Tying the cap to the supplier's insurance cover is a particularly clean approach, because it gives the customer a remedy the supplier can actually fund. ## The cap is only half the clause A cap works together with an exclusion of certain loss types. A typical clause first excludes indirect and consequential loss (lost profits, lost revenue, lost data, loss of goodwill) entirely, then caps whatever direct loss remains. This is crucial to understand: a high cap is worth little if the loss you fear is excluded altogether as consequential. When you assess a cap, always read the exclusion beside it. See our separate explainers on consequential damages waivers and on negotiating the limitation of liability clause. ## Carve-outs: what sits outside the cap Some liabilities are usually excluded from the cap, so they are either unlimited or subject to a separate, higher "super cap." Common carve-outs: - Death or personal injury caused by negligence, and fraud. In many jurisdictions these cannot lawfully be capped at all, so a cap that tries to is unenforceable to that extent.
• Data protection and security breaches. Often the biggest real-world risk, frequently given a super cap or left uncapped.
• Breach of confidentiality.
• IP infringement indemnities.
• A party's obligation to pay the fees it owes. The customer should still have to pay. Deciding which items are fully uncapped and which get a super cap is usually where most of the negotiation happens. ## Per-contract or aggregate? Where one master agreement sits above many orders or statements of work, the cap can apply per order or in aggregate across all of them. A per-order cap protects the customer better, because one order's cap is not used up by an unrelated order's problem. A single aggregate cap limits the supplier's total exposure. This should be settled deliberately, not left ambiguous. ## What the law allows Liability caps are generally enforceable between businesses, but there are limits that vary by country: - Many jurisdictions forbid capping or excluding liability for death or personal injury caused by negligence, and for fraud.
• In the UK, the Unfair Contract Terms Act 1977 requires that limitations in many contracts be reasonable, and a cap that is unreasonable can be struck down.
• In consumer contracts, statutory protections often override caps entirely.
• Some civil-law systems limit the ability to exclude liability for gross negligence. Because of this, a cap drafted for one country may not hold in another. Confirm enforceability in the governing-law jurisdiction before relying on the number. ## Common mistakes and misunderstandings - Focusing on the headline cap while ignoring the exclusion that removes the loss you actually fear.
• Assuming a cap protects against everything, when carve-outs and unenforceable items sit outside it.
• Accepting a cap of a few months' fees for a system whose failure could halt the business.
• Trying to cap liability for death, personal injury or fraud where the law forbids it.
• Forgetting to carve the paying party's payment obligation out of the cap.
• Leaving per-order versus aggregate unresolved under a master agreement. ## Seeing your caps across every contract One cap is easy to read. The portfolio is the hard part: which of your contracts have a cap at only a few months' fees, which carve data breach out, which are tied to insurance, and where you accepted a one-sided cap years ago. When contracts live in folders and inboxes, those answers take days to assemble. Keeping them in one connected workspace, where each agreement's cap and carve-outs are surfaced clause by clause and shown as favourable, neutral or unfavourable to you, turns scattered risk into something you can see at a glance. Adira is built around that idea, on frontier AI, so your caps and their carve-outs sit together. If you cannot quickly answer "which of our contracts have an uncapped data-breach liability?" that is the gap worth closing. More at Adira. The explanation above stands on its own. ## Bottom line A liability cap is the agreed ceiling on what one party can recover from the other, and it only makes sense read alongside the exclusion of consequential loss and the carve-outs that sit outside it. Set the cap at a level that matches your realistic worst case, tie it to insurance where you can, carve out the risks that can truly hurt you, and confirm the whole clause is enforceable in the governing-law country. For anything material, have a qualified lawyer review it.