Policy Interpretation Disputes: A Complete UK Guide

A policy interpretation dispute arises when an insurer and a policyholder disagree about what the words of an insurance policy actually mean and whether a particular loss is covered. Under English law these disputes are settled by reading the policy objectively — giving the words their ordinary meaning in the context of the whole contract — and a wrong rejection can often be reversed without going to court.

Most coverage arguments are not about whether something happened; they are about what the policy says about it. Did “flood” include a burst water main? Does a “communicable disease” extension respond to a national lockdown? Was a security condition a strict warranty or merely a risk-mitigation term? The answers turn on construction — the legal exercise of finding the meaning of contractual language.

This guide explains how UK courts construe insurance policies, the rules that favour policyholders (and their limits), the difference between conditions, warranties and conditions precedent, how the Insurance Act 2015 changed the law on warranties, and how the landmark COVID-19 business interruption test case reshaped the field. It is written for UK businesses and individuals in 2026 and sets out every realistic route to resolving a dispute, from a simple file review to litigation.

What a policy interpretation dispute is

An insurance policy is a contract. Like any contract, its words have to be given meaning, and when the insurer and the policyholder read those words differently the result is a policy interpretation dispute. These disputes are distinct from disputes about the facts (for example, whether a fire actually occurred or how much damage it caused). Here, the facts may be agreed; what is contested is the scope of cover the policy promises.

Typical flashpoints include: whether a particular peril falls within an insuring clause; whether an exclusion removes cover the insuring clause appears to grant; what a defined term such as “subsidence” or “occurrence” actually captures; and whether a condition the insured did not strictly comply with defeats the claim. Because policies are drafted by insurers and sold to a wide market, the same wording is litigated repeatedly, which is why a handful of Supreme Court decisions now govern the whole field.

Interpretation disputes sit close to, but are not the same as, broader liability disputes: the question is not who is at fault for a loss, but whether the policy bought to cover that loss actually responds to it.

How English law construes a policy

The modern law of contractual interpretation is set out in two Supreme Court decisions that every coverage lawyer cites. The exercise is objective: the court asks what a reasonable person, with all the background knowledge reasonably available to the parties at the time of contracting, would have understood the language to mean. The parties’ private intentions or hopes are irrelevant.

Arnold v Britton [2015] UKSC 36 — the primacy of the words

In Arnold v Britton the Supreme Court emphasised that the clearer the language, the harder it is to depart from its natural meaning. Lord Neuberger warned against using “commercial common sense” to rescue a party from a bad bargain: the fact that a clause produces a harsh or even disastrous outcome is not a reason to give it anything other than its ordinary meaning. Commercial common sense and surrounding context matter, but they cannot be invoked to rewrite words the parties chose. The court may only take into account facts that existed and were known or reasonably available to both parties when the contract was made.

Wood v Capita [2017] UKSC 24 — interpretation as a unitary, iterative exercise

In Wood v Capita Insurance Services Ltd the Supreme Court confirmed that Arnold had not pulled away from earlier authority such as Rainy Sky. Interpretation is a single, unitary exercise: the court reads the language and the relevant context together and tests rival meanings against the contract as a whole, its commercial consequences and business common sense — an iterative process. Where a clause is well drafted, the words carry more weight; where the drafting is loose or informal, context and commercial sense carry more. Textualism and contextualism are tools, not rival camps.

Reading the contract as a whole

No clause is read in isolation. A definition section, an insuring clause, conditions and exclusions all interact, and a meaning that makes one clause redundant or contradictory is usually rejected. Schedules, endorsements and the proposal form (where incorporated) form part of the contractual matrix and are read together.

Contra proferentem and its limits

The contra proferentem rule is the principle most policyholders have heard of: where a clause is genuinely ambiguous, it is construed against the party who drafted it — almost always the insurer. The rationale is to discourage insurers from using vague language to keep an escape route open, and to protect the policyholder who had no hand in the drafting.

But the rule has real limits, and policyholders who rely on it too heavily lose. Three points matter:

• It is a tie-breaker, not a starting point. The rule only applies after the ordinary principles of construction in Arnold and Wood have been exhausted and a genuine ambiguity remains. If the words have a clear natural meaning, contra proferentem never gets off the ground.
• Exclusions are read narrowly, but only if truly ambiguous. An exclusion that takes away cover the insuring clause grants is construed restrictively, so that an exclusion does not swallow the cover. Yet this only operates where there is real doubt; a clear exclusion is given full effect.
• Its force fades in negotiated wordings. In heavily negotiated commercial policies — where both sides had advisers and bargaining power — the courts give the rule a very limited role, because the wording reflects a freely chosen allocation of risk.

In the COVID-19 test case the Supreme Court applied these orthodox principles rather than reaching automatically for contra proferentem, a reminder that the rule is a last resort.

Conditions, warranties and conditions precedent

Whether a policyholder’s non-compliance defeats a claim depends entirely on how the relevant term is classified. The label the policy uses is not decisive; the court looks at substance and effect. The three main categories behave very differently.

Because the consequences are so different, classification is itself a frequent battleground. Insurers increasingly draft terms as conditions precedent to maximise their effect, so policyholders should check whether a clause genuinely qualifies as one or has merely been labelled that way.

The Insurance Act 2015 and warranties

The Insurance Act 2015, in force since 12 August 2016, reformed the harsh old law of warranties and remains central to interpretation disputes in 2026. Three reforms matter most.

Section 10 — breach of warranty is suspensory, not terminating

Under the old common law, breach of any warranty automatically and permanently discharged the insurer from all liability from the moment of breach, even if the breach was trivial, unconnected to the loss and later put right. Section 10 abolishes that rule. Breach now merely suspends the insurer’s liability for losses occurring during the period of breach; once the breach is remedied, cover revives for losses occurring afterwards. So a security alarm left off for a week defeats only a burglary that happens during that week — not one that happens the following month after the alarm is back on.

Section 11 — terms not relevant to the actual loss

Section 11 goes further. Where a term (whether a warranty or a condition precedent) is designed to reduce the risk of loss of a particular kind, at a particular location or at a particular time, the insurer cannot rely on the breach to refuse a claim if the policyholder shows that the non-compliance could not have increased the risk of the loss that actually occurred. A failure to maintain a fire alarm cannot be used to reject a flood claim. The first reported decision applying section 11 emerged in 2024, confirming that the courts will hold insurers to this causal connection.

Basis-of-the-contract clauses abolished

Section 9 abolishes “basis of the contract” clauses, which used to convert every statement on a proposal form into a warranty. These reforms cannot be contracted out of to the consumer’s disadvantage, and in non-consumer contracts any attempt to do so must satisfy strict transparency requirements. For most rejected claims involving a security or maintenance condition, the 2015 Act now gives the policyholder a far stronger position than the pre-2016 law.

Exclusions, ambiguity and definitions

Exclusions carve cover back out of the insuring clause. Because they take away what the insuring clause gives, they are construed narrowly where genuinely ambiguous, and the insurer must bring the loss squarely within the exclusion’s terms. Common disputed exclusions include those for “gradual” damage, “faulty workmanship”, “wear and tear” and various trade-specific carve-outs.

Definitions are the quiet engine of most coverage disputes. A policy will define key terms — “flood”, “storm”, “subsidence”, “occurrence”, “period of insurance” — and the dispute often collapses into whether the facts fall within the definition. Where a term is undefined, it takes its ordinary meaning to a reasonable reader, informed by any technical or trade usage that both parties would have understood.

• Ambiguity is not the same as disagreement. A clause is only ambiguous if it is reasonably capable of more than one meaning after ordinary construction — not merely because the parties argue about it.
• Specific wording beats general. A specifically negotiated endorsement or schedule entry usually prevails over inconsistent general printed conditions.
• Examples do not limit a general definition unless the policy says so; a list introduced by “including” is generally illustrative, not exhaustive.

Burden of proof: insuring clause vs exclusion

The allocation of the burden of proof frequently decides a marginal interpretation dispute. The rule is settled:

This split means that if the insurer rejects a claim by pointing to an exclusion, it carries the legal burden of establishing that the exclusion fits the facts. Where that is doubtful, the burden alone can tip the dispute in the policyholder’s favour.

Case study: the COVID-19 business interruption test case

The single best illustration of policy interpretation in action is The Financial Conduct Authority v Arch Insurance (UK) Ltd & Others [2021] UKSC 1. After the 2020 lockdowns, tens of thousands of SMEs claimed under business interruption (BI) extensions, and insurers refused on the basis that the wordings did not cover a national pandemic. The Financial Conduct Authority brought a test case to settle 21 representative wordings affecting around 370,000 policyholders and disputed sums of at least £1.2 billion.

What the Supreme Court decided

On 15 January 2021 the Supreme Court found substantially for policyholders across six issues:

• Disease clauses — wording covering business interruption caused by a notifiable disease “within” a stated radius generally responded to COVID-19, because the disease was treated as a single indivisible cause operating both inside and outside the radius.
• Prevention-of-access and hybrid clauses — were read more generously than insurers argued, capable of responding to restrictions even where premises were not fully closed.
• Causation — the court held that each individual case of COVID-19 was an effective cause of the government measures and the loss; the insured did not have to prove its loss would not have happened “but for” the local outbreak.
• Trends clauses — could not be used to strip out the very pandemic effects the cover was meant to capture, so adjustments must not reduce the indemnity to nil.
• Orient-Express overruled — the much-criticised Orient-Express Hotels decision, on which insurers relied to argue that wider catastrophe losses defeated cover, was overruled.

Why it matters for interpretation generally

The judgment is a masterclass in the Arnold/Wood method: the court read each wording on its own terms, in the context of the whole policy and its evident commercial purpose, and asked what the cover was plainly intended to do. It declined to let restrictive causation theory defeat cover the words granted, and it treated contra proferentem as a last resort. The practical effect was that many thousands of valid claims had to be paid — a powerful reminder that a confident insurer rejection is not the final word on what a policy means.

How disputes get resolved

Litigation is the last resort, not the first. A policy interpretation dispute can be resolved at several escalating stages, and most are settled well before a courtroom.

1. Coverage review and reasoned challenge

The first step is a careful read of the full policy — schedule, wording, endorsements and definitions — against the facts and the insurer’s rejection letter. A reasoned letter that sets out the natural meaning of the relevant clauses, the burden of proof and the relevant authorities frequently prompts an insurer to reconsider, especially where it has relied on an exclusion it cannot make fit.

2. Internal complaint

If the insurer maintains its position, the policyholder uses the insurer’s formal complaints procedure. The insurer must respond within eight weeks. Sound claims-handling practice standards apply, and a poorly reasoned rejection often weakens at this stage.

3. Financial Ombudsman Service

For eligible complainants — consumers and most small businesses (micro-enterprises and small businesses under the FCA thresholds) — the Financial Ombudsman Service (FOS) offers a free, binding determination. The FOS decides what is fair and reasonable in the circumstances, which can be broader than a strict legal reading, and its decisions bind the insurer up to the FOS award limit (£430,000 for complaints about acts from April 2019, reviewed periodically). A complaint must usually be referred within six months of the insurer’s final response.

4. Expert opinion and mediation

Where the dispute turns on a genuinely arguable point of construction, an early opinion from experienced coverage counsel can crystallise the strength of each side’s case and drive settlement. Mediation — a confidential, without-prejudice negotiation led by a neutral mediator — resolves many coverage disputes at a fraction of the cost of litigation, and the courts expect parties to consider it.

5. Litigation

If all else fails, the dispute is decided by the courts (in England and Wales, usually the Commercial Court or the County Court depending on value). Litigation is a reserved activity: it must be conducted by a regulated solicitor or barrister. The advantage of a court judgment is a binding, enforceable interpretation; the disadvantages are cost, delay and exposure to the other side’s costs if you lose.

Common mistakes to avoid

• Treating the rejection letter as final. A confident insurer “no” is an opening position, not a verdict on what the policy means.
• Arguing fairness instead of wording. Post-Arnold, “it would be unfair if I weren’t covered” is not a winning argument; the natural meaning of the words is.
• Overplaying contra proferentem. It only applies to genuine ambiguity after ordinary construction fails, and barely at all in negotiated wordings.
• Ignoring the burden of proof. If the insurer relies on an exclusion, it must prove it — don’t concede that burden by trying to disprove the exclusion yourself.
• Assuming a warranty breach is fatal. Since the Insurance Act 2015, breach merely suspends cover, and an irrelevant breach cannot defeat a claim at all (s.11).
• Missing FOS time limits. Eligible complainants generally have six months from the insurer’s final response to go to the Ombudsman.
• Reading clauses in isolation. Definitions, conditions and exclusions interact; a meaning that ignores the rest of the policy usually fails.

Frequently asked questions

This article is for general information only and does not constitute legal or accountancy advice. Hayhills Limited, trading as Hayhills Legal Advisory, provides non-reserved legal advisory services. Always check current requirements at GOV.UK.