FCA authorised · FRN 7249520117 325 0027Quote & buy →
Reviewed by Matthew Bartlett, Director · Last reviewed 2026-06-23
"Aggregation" is the policy mechanism by which separate claims are treated as a single claim against a single policy limit. UK case law on aggregation has developed mostly through PI disputes. This entry reviews the leading authorities and the practical effect for UK professionals in 2026.
A PI policy typically pays up to its limit "any one claim" or "in the aggregate". Where multiple claims could arguably be one claim under aggregation, the difference is enormous: ten £500k claims may be ten separate £500k payouts (separate claims) or one £500k payout (aggregated). The wording of the aggregation clause and the case law applying it determine the answer.
Aggregation cuts both ways, which is why it is so often litigated. For the policyholder, aggregating many small claims into one can be catastrophic where the limit is modest: a single limit is exhausted and every subsequent claimant looks to the firm's own balance sheet. In other situations aggregation helps the insured, because it means only one self-insured excess (deductible) is deducted rather than one per claim. The same wording that protects the insurer against a flood of separate limits can protect the insured against a stack of separate excesses. Reading an aggregation clause therefore means asking two questions at once: how many limits are engaged, and how many excesses are payable.
Aggregation language is rarely bespoke. Most UK PI wordings use one of a handful of standard formulations, and the courts interpret those formulations against the ordinary meaning of the words rather than any general "policy" of insurance. That is why a change of a few words — "arising from one originating cause" versus "arising from one event" versus the Solicitors Regulation Authority (SRA) "series of related matters or transactions" — can move millions of pounds of exposure between insurer and insured. The clause is read as a commercial bargain, and the burden of showing that claims should be treated as one usually falls on the party that benefits from that treatment on the facts.
The Supreme Court considered the SRA Minimum Terms aggregation clause in the context of failed property development schemes. The clause aggregated claims arising from "similar acts or omissions in a series of related matters or transactions". The court held:
The result: solicitors who had advised on multiple unrelated property transactions using a flawed methodology faced individual claims, not aggregated ones. The single development scheme losses aggregated; the cross-scheme losses did not.
Two points from the judgment are worth holding onto at renewal. First, the Supreme Court declined to read the words "related matters or transactions" as requiring the transactions to be part of a single overarching transaction; it was enough that they were in some way dependent on one another — that they fitted together in a real, unifying way rather than merely resembling one another. Second, the court treated the search for that unifying factor as intensely fact-sensitive. Whether a group of matters aggregates is decided on the actual connections between them, not on a label. For a firm buying cover, that means the safe assumption is uncertainty: you cannot know in advance exactly how a future group of claims will fall, so the limit must be sized for the realistic worst case rather than the best.
The House of Lords (now Supreme Court) considered aggregation of mis-selling claims. The court held that "originating cause" aggregation language is broader than "event" aggregation. Originating cause looks back to the root; event aggregation looks at the proximate cause. The same individual claims can aggregate under originating-cause wording and not aggregate under event wording.
The practical lesson is that "cause" and "event" are not interchangeable, even though they read similarly. An "event" is something that happens at a particular time and place; a "cause" can be a continuing state of affairs — a defective training programme, a flawed template, a systemic compliance failure — that never happens at a single moment at all. A book of mis-selling complaints driven by one bad process will often share an originating cause but no single event, so the choice of wording effectively decides whether they collapse into one claim.
Because aggregation controls how many times the limit can be recovered, it should drive the way you size cover — not the headline number alone. Two firms of identical fee income can need very different limits depending on how their work aggregates. A practice that does many small, unrelated matters is exposed to a large number of separate claims, so it benefits from either an "any one claim" basis or a generous aggregate with reinstatement. A practice concentrated on a handful of large, interconnected schemes faces fewer but larger events, so a single well-judged limit may suffice.
Three levers interact with aggregation when sizing cover: the limit basis ("any one claim" versus "in the aggregate"), the availability and cost of a limit reinstatement, and the excess structure. Where claims are likely to aggregate, an aggregate limit with no reinstatement is the most fragile arrangement, because one linked group of matters can consume the whole policy for the year. Where claims are likely to be treated separately, the excess becomes the pressure point, because each separate claim carries its own deductible. A broker's job is to model your realistic worst case against the wording that will actually apply, and to make sure the limit is set above the largest plausible aggregated loss rather than the average claim. Fair presentation of the risk under the Insurance Act 2015 also matters here: disclosing the concentration and interconnection of your work helps ensure the policy responds as intended.
UK PI policies use three main aggregation patterns:
The aggregation clause in the primary policy may differ from the excess layer's aggregation clause. A primary that aggregates broadly and an excess that aggregates narrowly creates a "gap" — claims that aggregate at primary level (exhausting it on one bill) may NOT aggregate at excess level (engaging multiple sub-limits). Always check parity at renewal.
The reverse mismatch is just as dangerous. If the excess layer aggregates more broadly than the primary, a group of claims might be treated as separate at primary level (so several excesses and limits are chewed through) yet as one at excess level (so the top cover is capped as a single claim). Layered PI programmes should therefore be read as one instrument, with the aggregation, notification and claims-definition wording checked for consistency from the ground up. "Follow-form" excess wordings help, but only if the follow really does mirror the primary on aggregation.
A worked scenario shows why this is not academic. Suppose an architects' practice designs the same balcony detail across a 30-unit development, and the detail proves defective in every unit. Under Woodman-style reasoning those matters are connected within one scheme, so they are likely to aggregate: one limit responds to the whole development. Now suppose the same practice reused a similar (but not identical) detail on ten unconnected projects for ten unrelated clients. Those are far more likely to be treated as separate claims, so ten excesses fall due and ten limits are potentially engaged. Same firm, same underlying error, radically different insurance outcome — decided entirely by the connection between the matters and the aggregation wording.
It depends on the limit and the excess. Aggregating claims into one means only one excess is payable, which helps; but it also means only one limit responds, which hurts if the aggregated loss is large. The right answer is set by matching the wording to how your work actually clusters.
Following Lloyd's TSB, "originating cause" looks back to the root of the losses and is the broader test, so more claims collapse into one. "Event" aggregation looks at a single triggering happening and is narrower. The same set of claims can aggregate under one and not the other.
Usually not, on the SRA Minimum Terms wording considered in Woodman. A repeated method applied independently to unrelated clients tends to produce separate claims. Aggregation generally needs a real connection between the matters themselves, not just a shared error.
Set the limit above your largest plausible aggregated loss, not your average claim. If your work concentrates on large interconnected schemes, one big event can consume an aggregate limit, so a higher limit or a reinstatement is often prudent. A broker can model this against your actual wording.
Because the excess layer can define aggregation differently from the primary. If the two do not match, claims can behave one way at primary level and another at excess level, leaving a gap in cover. We check parity across every layer at renewal.
Apex Insurance Brokers Limited reads aggregation wording line by line for UK PI placements. FCA firm reference number 724952. The Woodman case law makes the wording particularly important for any profession dealing with multiple unrelated clients on similar matters — solicitors, accountants, IFAs.
Apex Insurance Brokers serves UK professional services firms and commercial businesses. Call 0117 325 0027, email info@apexinsurancebrokers.co.uk, or request a quotation.
Get a quote