Limitation of liability clauses are the most important commercial term in most UK service agreements. They determine the maximum financial exposure each party has if something goes wrong — a software outage, a data breach, a missed deadline, a regulatory failure. A well-drafted limitation clause balances protecting the supplier from catastrophic exposure with giving the customer meaningful recourse. A badly drafted one shifts almost all risk to one party. UK law imposes specific constraints on limitation clauses. The Unfair Contract Terms Act 1977 (UCTA) controls liability exclusions in B2B contracts via the reasonableness test, and the Consumer Rights Act 2015 governs fairness in B2C contracts. Certain liabilities cannot be excluded at all — death or personal injury caused by negligence, fraud, and breach of statutory implied terms in consumer contracts. Beyond these, the contract terms govern.
What is a Limitation of Liability?
A limitation of liability clause caps and excludes a party's potential liability under the contract. Typical structures include: (1) a financial cap (often expressed as a multiple of fees paid, e.g., 1x annual fees); (2) exclusion of indirect, consequential, and special losses; (3) carve-outs for specific high-risk categories (IP infringement, data breach, breach of confidentiality); and (4) unbreakable carve-outs for liabilities that cannot be excluded by law (death, personal injury, fraud). Under UCTA, the reasonableness test (s 11 and Schedule 2) governs whether a B2B exclusion or limitation is enforceable, considering bargaining power, choice, knowledge, and insurance availability. In B2C contracts, the Consumer Rights Act 2015 applies a fairness test.
Red flags to watch for
A liability cap that bears no rational relationship to the value of the services or the magnitude of potential harm may fail the UCTA reasonableness test in B2B. For a £1m contract, a 0.5x cap means the supplier's maximum exposure is £500k for any conceivable failure — including catastrophic data breaches affecting thousands of customers. Caps of 1-2x annual fees are common; significantly lower caps should be challenged.
Standard practice excludes indirect/consequential losses, but excluding all loss of profit (even direct loss of profit caused by clear breach) can be unreasonable. The case law (Sugar Hut Group Ltd v A J Insurance Service [2016] EWCA Civ 46) makes clear that 'loss of profit' is often a direct loss in a commercial context and should not be excluded as 'indirect'.
Specific high-risk categories — IP infringement indemnity, data breach indemnity, breach of confidentiality — typically should sit outside the general cap or have a higher specific cap. If everything is bundled under a single low cap, the customer has no meaningful recourse for the most serious failures.
Liability for death or personal injury caused by negligence cannot be excluded under UCTA s 2(1). Liability for fraud cannot be excluded as a matter of public policy. The Misrepresentation Act 1967 controls exclusions of fraudulent misrepresentation. A limitation clause missing these carve-outs may be unenforceable in those respects, and indicates poor drafting overall.
A contractual time-bar shorter than the statutory limitation period (6 years for simple contracts under the Limitation Act 1980) is permissible in B2B but must be reasonable. Time-bars of 3 months or less are often unreasonable, particularly for latent defects or claims requiring investigation.
Some agreements treat service credits as reducing the cap. If your only meaningful remedy for downtime is service credits, and those credits also count against the cap, you have effectively no cap-protected recourse for serious failures.
Sophisticated B2B negotiations often produce symmetric caps. A clause that limits the supplier's liability to fees paid while imposing unlimited liability on the customer is asymmetric and often unjustifiable.
Your legal rights
UK limitation clauses are governed by the Unfair Contract Terms Act 1977 (UCTA) for B2B contracts and the Consumer Rights Act 2015 (CRA) for B2C contracts. UCTA s 2 controls exclusion of liability for negligence; s 3 controls liability in standard form contracts. The reasonableness test under UCTA s 11 and Schedule 2 considers bargaining power, customer choice, customer knowledge, and insurance. CRA s 31, 47, and 57 control terms in consumer goods, digital, and services contracts respectively. Liability for death or personal injury caused by negligence cannot be excluded. Liability for fraud cannot be excluded. The Limitation Act 1980 sets the default limitation period (6 years for simple contracts, 12 for deeds) which contracts may shorten only within reason. Case law including AstraZeneca UK Ltd v Albemarle International Corporation [2011] EWHC 1574 (Comm) and Sugar Hut Group Ltd v A J Insurance Service [2016] EWCA Civ 46 governs interpretation of indirect/consequential loss exclusions.
Questions to ask before you sign
- 1What is the cap, and how does it relate to the potential magnitude of harm — not just the fees paid?
- 2Are direct losses (including direct loss of profit) recoverable, or are they swept into 'indirect losses' and excluded?
- 3Are high-risk categories — IP infringement, data breach, confidentiality breach — carved out from the general cap?
- 4Are death/personal injury caused by negligence, fraud, and fraudulent misrepresentation properly carved out?
- 5What is the contractual time-bar for bringing claims, and is it reasonable for the type of claims that could arise?
- 6Do service credits reduce the cap, or are they in addition to other remedies?
- 7Is the cap symmetric between the parties, and if not, why?
Disclaimer: This guide is for educational purposes only and does not constitute legal advice. Contract law varies by jurisdiction and individual circumstances. Always consult a qualified legal professional before making decisions based on this information.