Limitation of Liability in Technology Contracts: What Businesses Need to Know

Key takeaways

• Words decide outcomes: English courts generally hold commercial parties to the risk allocation they agreed. Recent technology cases show that small drafting differences can change recoverable damages by millions of pounds.
• Three moving parts: Most clauses exclude certain types of loss, cap what remains, and carve out liabilities that sit outside the cap. Each needs to be drafted, and read, with the others.
• Statutory limits: Some liabilities cannot be excluded at all, and caps in standard terms must be reasonable under the Unfair Contract Terms Act 1977.
• Align with insurance: A cap that bears no relationship to the fees, the risks or the insurance available is harder to justify and may leave liability uninsured.

When a technology project fails, a SaaS platform goes down or customer data is compromised, the limitation of liability clause often determines how much can be recovered. It is one of the most negotiated provisions in any technology contract, and one of the most frequently litigated. Yet it is often treated as boilerplate until something goes wrong.

This article explains how limitation of liability clauses work in technology contracts under English law, the statutory limits on them, and what recent case law says about drafting them. It is relevant to suppliers seeking to manage their exposure and to customers seeking to ensure they have a meaningful remedy.

How does a limitation of liability clause work?

Most clauses have three parts:

• Exclusions: categories of loss that cannot be recovered at all, typically indirect or consequential loss, and often loss of profits, revenue, savings, goodwill or data.
• The cap: the maximum amount recoverable for losses that are not excluded, usually a fixed sum, a multiple or percentage of the charges, or the greater of the two.
• Carve-outs: liabilities that are unlimited or subject to a different cap, either because the law requires it or because the parties have agreed it, such as liability for breach of confidentiality, data protection or IP indemnities.

What liability cannot be excluded or limited?

Under the Unfair Contract Terms Act 1977 (UCTA), a business cannot exclude or restrict liability for death or personal injury resulting from negligence. Liability for other loss or damage caused by negligence can be excluded or restricted only if the term is reasonable. Where one party deals on the other’s written standard terms of business, the party relying on the terms cannot exclude or restrict its liability for breach of contract unless the term is reasonable. A party also cannot exclude liability for its own fraud.

Reasonableness is judged by reference to the circumstances known or contemplated when the contract was made, and the party relying on the clause must show that it is reasonable. Where liability is limited to a specified sum, the court will consider the resources available to the party relying on the limit and how far it could have covered itself by insurance. In consumer contracts, the Consumer Rights Act 2015 applies instead and imposes stricter controls.

In practice, many technology contracts are negotiated between sophisticated businesses with legal advice, and courts are generally reluctant to interfere with the risk allocation they have agreed. Where a supplier contracts on its standard terms, however, its cap should be justifiable by reference to the fees, the risks and the insurance it holds.

How should the cap be structured?

Fixed sum or link to the charges?

Caps are commonly set at a fixed amount, a percentage of the charges paid or payable in a period (often the previous twelve months), or the greater of the two. Linking the cap to the charges keeps it proportionate, but needs care. A cap based on charges in the previous twelve months may be very low if a claim arises early in the contract, so a minimum figure for the first year is often included. “Payable” should be defined, since it may refer to amounts already invoiced or to amounts expected to become due.

One cap, or several?

The parties should state clearly whether the cap applies to all claims in aggregate, to each claim or series of connected claims, or to each contract year. The difference can be very significant. In Drax Energy Solutions Ltd v Wipro Ltd [2023] EWHC 1342 (TCC), a supplier of a failed IT system relied on a cap of 150% of the charges paid or payable in the twelve months before the claim arose. The customer claimed around £31 million and argued that a separate cap applied to each of its claims. The court held that the clause created a single aggregate cap of about £11.5 million for all claims. By contrast, in Royal Devon and Exeter NHS Foundation Trust v ATOS IT Services UK Ltd [2017] EWCA Civ 2196, the Court of Appeal held that a clause provided for two separate caps, one for claims arising in the first contract year and another for later claims.

Before or after set-off?

Where both parties have claims, the order in which the cap and any set-off are applied matters. In Topalsson GmbH v Rolls-Royce Motor Cars Ltd [2024] EWCA Civ 1330, a software supplier was liable to its customer for around €8 million in termination damages, and was itself owed around €0.8 million, under a contract capping each party’s total liability at €5 million. The Court of Appeal held that the cap applied to each party’s liability before set-off, so that the supplier paid around €4.2 million rather than the full €5 million. It also held that contractual interest on late payment fell outside the cap.

Which losses can be excluded?

Indirect and consequential loss

In English law, “indirect” or “consequential” loss is generally understood to mean loss falling within the second limb of Hadley v Baxendale: loss that does not arise naturally from the breach but only from special circumstances known to the parties. It does not exclude all losses that follow from a breach. Businesses that intend to exclude particular losses, such as loss of profits, should therefore say so expressly.

Loss of profits and wasted expenditure

Exclusions of particular types of loss will be interpreted according to their natural meaning, and the consequences can cut either way. In Soteria Insurance Ltd v IBM United Kingdom Ltd [2022] EWCA Civ 440, an IT supplier relied on an exclusion of loss of profits, revenue and savings. The Court of Appeal held that this did not exclude the customer’s claim for expenditure wasted on a failed project, which was a different type of loss. In EE Ltd v Virgin Mobile Telecoms Ltd [2025] EWCA Civ 70, by contrast, the Court of Appeal held by a majority that an exclusion of liability for “anticipated profits” prevented a claim of around £25 million for charges the claimant said it would have received but for the defendant’s breach of an exclusivity obligation.

These cases show the importance of thinking about which losses are likely to be the main consequence of a breach. In a technology contract, loss of profits, wasted expenditure, the cost of procuring a replacement system and the cost of restoring data may be the customer’s principal losses, and excluding them may leave little to which the cap applies.

What should sit outside the cap?

Common carve-outs and separate caps include:

• Mandatory exclusions. Death or personal injury caused by negligence, fraud and any other liability that cannot lawfully be limited.
• IP indemnities. Claims by third parties alleging that the supplier’s technology infringes their rights. Suppliers often resist uncapped indemnities, particularly for AI products, where the law on infringement is still developing.
• Data protection and security. Increasingly dealt with through a separate, higher cap rather than unlimited liability.
• Confidentiality. Often uncapped or subject to a higher cap.
• Charges. The customer’s obligation to pay the charges is normally outside the cap.
• Deliberate or wilful default. Sometimes uncapped, but the terms should be defined, since their meaning is not settled.

Carve-outs need to be drafted precisely. In Triple Point Technology Inc v PTT Public Company Ltd [2021] UKSC 29, a software contract excluded liability for “negligence” from its cap. The Supreme Court held that this extended to negligent breach of the supplier’s contractual duty of care, not only to the tort of negligence, so that the customer’s claim for damages was not capped. The same case confirmed that liquidated damages for delay accrue up to termination, unless the contract clearly provides otherwise.

Service credits and other remedies

SaaS and managed services agreements often provide for service credits if service levels are not met. The agreement should say whether service credits are the customer’s sole remedy for the relevant failure or are without prejudice to a claim for damages, whether they count towards the cap, and at what point persistent failure gives rise to a right to terminate. Where the contract allows for liquidated damages for delay, their interaction with the cap and with termination should also be addressed.

Insurance

The cap should be considered alongside the supplier’s insurance. Where the cap exceeds the insurance limit, or where the contract includes indemnities that insurers will not cover, part of the liability may be uninsured. Where the cap applies per claim but the insurance has an aggregate limit, several claims may exhaust the cover. The level of insurance available is also relevant to whether a cap is reasonable under UCTA.

Limitation of liability: a practical checklist

When drafting or negotiating limitation of liability in a technology contract, it is advisable to:

• Identify the likely losses. Consider what each party would actually lose if the technology failed, and make sure the clause works for those losses.
• State how the cap operates. Say expressly whether it applies in aggregate, per claim or per year, and how it is calculated, including in the first year.
• Define excluded losses. Do not rely on “indirect or consequential loss” alone, and consider whether excluding loss of profits or wasted expenditure would leave a meaningful remedy.
• Draft carve-outs precisely. Define terms such as negligence, wilful default and data protection breach, and set separate caps where appropriate.
• Address set-off and interest. Say whether the cap applies before or after set-off, and whether interest and service credits count towards it.
• Check consistency. Ensure the clause works with indemnities, service levels, liquidated damages and termination provisions.
• Align with insurance. Check the cap and carve-outs against the insurance in place.

Where a project has already gone wrong, the limitation clause will be central to assessing the claim, as we discuss in our article on managing contractual risk when technology projects go wrong.

At Maybrook Law, we advise technology suppliers and customers on the drafting and negotiation of technology contracts, including limitation of liability, indemnities and service levels, and we advise on disputes when technology projects and services fail. If you are negotiating a technology contract, or would like to discuss any of the issues raised in this article, we would be pleased to advise on your position.

This article reflects the law as at . Our insights, articles and guides do not, and are not intended to, constitute legal advice or be an exhaustive review of all legal developments. Although every effort is made to ensure that the information provided in this article is accurate as of the publication date, please be aware that this area of law is developing rapidly and may be subject to change. Please seek legal advice before applying the information provided to any specific circumstances, transactions or legal issues.