Higher PI limits for architects and design firms

Reviewed by Matthew Bartlett, Director, Apex Insurance Brokers Limited · Last reviewed 2026-08-10

Why do architects and design consultancies outgrow standard limits?

Most design practices start with a PI limit set years earlier, often by reference to an appointment that has long since completed. The business then changes around it. Project values rise, the practice takes on lead consultant or principal designer roles, and the contractual matrix — appointments, warranties, third-party rights — grows more demanding. The limit that was comfortable at incorporation can look thin against the schemes now on the drawing board.

The ARB requires architects to hold adequate and appropriate professional indemnity insurance, and professional body requirements point the same way: adequacy is judged against the work you actually do, not the work you did when the policy was first placed. For a practice moving into larger residential schemes, higher-risk buildings, or design roles on major infrastructure, "adequate" is a moving target. The honest test is not what the market average buys, but what a single serious claim on your most demanding current project could cost — including defence costs, which on construction disputes routinely run to seven figures before any question of damages is settled.

There is a second, quieter driver: claims inflation. Rectification costs track construction cost inflation, and the cost of remedying a design defect discovered after practical completion — with the building occupied and access constrained — bears little relation to the original fee. A limit that once represented several multiples of your largest fee may now be exhausted by one remediation scheme.

What do collateral warranties do to the limit I actually need?

Every collateral warranty or third-party rights schedule a practice signs widens the population of parties who can bring a claim on the same design. A funder, a purchaser, and successive tenants may each hold rights in respect of one building; on a mixed-use scheme the warranty schedule can run to dozens of beneficiaries. The design error is singular, but the claims need not be — and where a policy operates on an aggregate basis, several claimants can be eroding the same limit simultaneously.

Warranties also commonly commit the firm to maintain a specified PI limit for a fixed period after completion, often twelve years where the underlying appointment is executed as a deed. That converts the limit from an annual purchasing decision into a long-term contractual obligation, one which survives changes in the insurance market and in the practice's own fortunes. Before signing, it is worth checking that the promised limit, the basis of cover and the maintenance period are all achievable — not just this year, but plausibly across the whole commitment. We cover the interaction in more detail in our guide to collateral warranties and PI insurance.

How has the Building Safety Act changed the liability tail?

The Building Safety Act materially lengthened the period over which certain claims relating to defective or unsafe buildings can be brought, in some respects retrospectively. In general terms, design work that would once have been comfortably time-barred may now be actionable, and work completed today carries a substantially longer prospective tail than the pre-Act position. For architects and fire engineers in particular, this changes the arithmetic of limit adequacy: the question is no longer only whether the limit is sufficient for current projects, but whether the programme can respond to claims emerging from work done many years earlier, judged against remediation costs at the date the claim arrives.

The Act's wider regime for higher-risk buildings has also raised the standard of what design duty-holders are expected to demonstrate. Practices operating in that space face both a longer exposure period and a more demanding compliance environment, and underwriters have responded accordingly: many wordings now carry fire safety restrictions, aggregate limits for certain perils, or specific conditions around cladding and external wall systems. A higher headline limit that sits behind a heavily restricted wording may deliver less protection than it appears to. Limit and wording have to be assessed together, and that is precisely where broker judgement earns its keep.

What if a client or framework demands more cover than we carry?

Framework agreements, public sector appointments and institutional developer requirements increasingly specify PI limits well above what a mid-sized practice has historically carried — sometimes with prescriptive requirements on the basis of cover as well as the amount. The commercial reality is that the limit requirement is often a gateway: without evidence of compliant cover, the practice cannot be appointed at all.

The instinct is simply to buy the number demanded. The better approach is to interrogate the requirement first. Is the limit required any one claim or in the aggregate? Does it apply to all liability or can fire safety or pollution sit on a different basis? Is there scope to negotiate the figure by reference to your actual scope of services — a practice providing concept design only carries a different exposure from the lead designer with site inspection duties? Requirements that look immovable are frequently negotiable once the client's real concern is understood, and where they are not, the additional limit can usually be structured efficiently rather than bought as a single expensive block.

Does the basis of cover matter as much as the headline limit?

For design firms, frequently more. A limit written any one claim reinstates in full for each separate claim; an aggregate limit is a single pot for the whole policy year, shared across every claimant and every project. Architects' PI has, in recent market conditions, often been written wholly or partly in the aggregate — particularly for fire safety exposures — which means a practice comparing its limit against a contractual requirement expressed "each and every claim" may not actually comply, whatever the number on the schedule says.

Aggregate structures can be softened: many wordings offer one or more reinstatements, and costs can sit in addition to rather than within the limit. Each variation changes what the programme will actually pay in a bad year. When assessing whether your limit is genuinely adequate, the basis, the treatment of defence costs, and any inner limits for specific perils are as material as the headline figure itself.

How are higher limits for design firms actually placed?

Beyond a certain point, a single insurer will rarely write the whole limit for a design practice — appetite for construction professional risks is finite, and insurers manage their exposure to any one firm carefully. Higher limits are therefore typically built as a programme: a primary layer, on which the wording and claims handling are anchored, with one or more excess layers sitting above it, each attaching when the layer below is exhausted.

Layered placements bring their own technical questions. Excess layers are usually placed on a follow-form basis, but "usually" is doing work in that sentence: any divergence between layers — on fire safety exclusions, aggregation language, or the basis of cover — creates gaps precisely where a large claim would find them. The sequencing matters too. A well-run placement secures the primary first, on the strongest available wording, and then builds the tower on terms consistent with it, rather than assembling capacity opportunistically and reconciling the wordings afterwards.

Expect underwriters on higher-limit design risks to want genuine detail: a breakdown of fees by discipline and project type, the proportion of work on higher-risk or residential buildings, cladding and external wall involvement past and present, the firm's approach to appointment review and liability caps, sub-consultant management, and full claims history with narrative on anything significant. Practices that present this well — and can show how lessons from past notifications changed procedure — consistently achieve better terms than the raw claims record alone would suggest. The submission is, in effect, the practice's advocacy document, and it deserves the same care as a competition entry.

How should we decide what limit to buy?

There is no formula, but there is a disciplined way to reason about it. In our experience the assessment should weigh:

• the realistic worst-case rectification cost on your largest and most complex current projects, not the average;
• the aggregate effect of warranties and third-party rights across live and recently completed schemes;
• the lengthened claims tail on building safety exposures, judged against future remediation costs;
• contractual and framework minimums you are committed to maintaining, and for how long; and
• what the practice could survive uninsured — the excess, any inner limits, and anything above the tower.

The output of that exercise is rarely "keep the limit we have". Sometimes it justifies a substantially higher tower; sometimes it shows the money is better spent improving the basis of cover or removing a restriction than adding another layer. Either way, the decision becomes evidence-based — which is also, not incidentally, the strongest demonstration that your insurance is adequate and appropriate for the work you do.

For practices whose requirements have moved beyond the packaged market, this is specialist territory: a modest number of insurers, placement sequencing that matters, and wordings where the detail decides outcomes. It rewards a broker who works at those limits routinely.

Apex Insurance Brokers Limited is authorised and regulated by the Financial Conduct Authority (FRN 724952). This article is general information, not advice on a specific policy.