The mental model
A professional sells advice or skill. When that advice or skill turns out to be defective, somebody who relied on it loses money or gets hurt. Professional liability is the policy that responds.
General liability protects against bodily injury and property damage caused by ordinary operations. Professional liability protects against the consequences of getting the professional work wrong. An architect designs a building that fails inspection. A consultant gives advice that costs the client a contract. A software vendor ships a release that crashes a customer's production. A lawyer misses a statute of limitations. None of these losses look like a slip and fall, and none of them sit on a CGL form. They are economic losses arising from the rendering of professional services, and they get their own product, their own legal framework, and their own claims handling culture.
E&O is about being wrong, not being unsafe. The trigger is a wrongful act in the rendering of professional services, not a physical event. The damages are usually pure economic loss. The policy is almost always claims-made. The defense costs are usually a meaningful share of the total exposure. If you internalize those four points you have the line.
Claims-made mechanics
An occurrence policy responds to losses that happen during the policy period, no matter when the claim is made. A claims-made policy responds to claims that are first made during the policy period, no matter when the underlying act occurred. Most professional liability is claims-made for two reasons: (1) the time between a wrongful act and a discovered loss is often years, sometimes decades, and occurrence pricing for that tail is brutal; (2) carriers want the ability to reprice annually as the loss environment changes, which they cannot do if they are still on risk for ten-year-old policies.
The four conditions for coverage
A claim under a claims-made policy generally needs four boxes ticked:
- The wrongful act occurred after the retroactive date.
- The wrongful act occurred before the end of the policy period.
- The claim was first made against the insured during the policy period (or any extended reporting period).
- The claim was reported to the carrier in accordance with the notice provisions.
Miss any one and there is no coverage. The retro date and the reporting requirement are the two places where coverage gets lost most often.
Claims-made and reported
Some forms are "claims-made" (the claim must be first made during the period, but reporting can extend a bit). Others are "claims-made and reported" (the claim must be both first made and reported during the period, with at most a short post-expiration reporting window). The distinction matters for renewal timing and for late notice arguments. Read the insuring agreement carefully.
An insured switches carriers at renewal. The new carrier issues a policy with a retro date matching the inception of the new policy, not the original retro date. Six months later a claim is reported for an act that occurred under the prior carrier's policy. The new policy excludes it (act before retro date). The prior policy excludes it (claim made after expiration, no ERP purchased). The insured has a coverage gap they did not see coming. The fix is matching retro dates at renewal or buying tail on the prior policy. Both fixes require somebody to read the binder carefully.
Retro dates and prior acts
The retroactive date is the earliest date of a wrongful act that the policy will cover. Acts before the retro date are excluded. Acts after the retro date that result in a claim during the policy period are covered.
The three retro-date scenarios
Prior acts coverage
"Prior acts" is industry shorthand for the period between the retro date and policy inception. Continuous prior acts means the carrier is covering everything from the retro date forward, including work the carrier never reviewed at the time it was performed. That is a lot of unknown exposure to assume. Carriers price prior acts as a separate factor on the rate, and routinely ask for a warranty statement that the insured is not aware of any circumstance that could give rise to a claim as of the policy effective date. False answers on that warranty are how carriers rescind.
The renewal application warranty
Most renewal applications include language like: "No partner, principal, or member is aware of any act, error, omission or circumstance that could reasonably give rise to a claim under the policy being applied for." If the insured signs that warranty knowing about a circumstance and does not disclose it, the carrier has grounds to deny coverage when the claim materializes. This is the single most-litigated coverage issue in E&O. Underwriters take the warranty seriously. Insureds, sometimes, do not.
ERPs and tail coverage
An Extended Reporting Period (ERP), also called "tail," extends the time during which the insured can report a claim under the expiring policy after that policy ends. It does not extend the policy period itself. An act that occurs after the policy period is still excluded. Only the reporting window is extended.
BERP, automatic, and supplemental
- Basic ERP (BERP). A short automatic extension (often 30 to 60 days) included in most claims-made forms at no additional premium. Useful for claims discovered immediately after expiration, but rarely sufficient for the long tail.
- Automatic ERP. Triggered by certain events (e.g., the carrier non-renews). Often free or low-cost. Term varies, often 60 days to 12 months.
- Supplemental ERP. An optional extended tail purchased at expiration, typically priced as a multiple of the expiring premium. One-year tails run roughly 100% of expiring premium, three-year tails 150-200%, six-year tails 200-250%, unlimited tails are negotiable. The exact factors are in the policy.
When tail matters most
Tail is critical at three moments: (1) when an insured retires or sells their practice and will no longer have an active policy to provide ongoing coverage; (2) when a firm dissolves and individual partners need to protect themselves for past acts; (3) when an insured switches carriers and cannot get matching retro dates from the new carrier. Brokers who sell tail well are doing real risk management. Brokers who forget to talk about tail are creating future E&O claims against themselves.
Many lawyer's, doctor's, and accountant's policies include a free unlimited tail that triggers if the insured dies, becomes permanently disabled, or retires after a stated period of continuous coverage (often five or ten years). It is a meaningful retention tool for carriers and a meaningful protection for individual professionals. The exact triggers and qualifying conditions vary across forms, and the tail is only available if certain conditions are met (typically retiring from the practice of the insured profession entirely, not just from the firm).
Defense within or outside limits
One of the most consequential structural choices in E&O is whether defense costs erode the policy limit or sit outside of it. This decision drives both pricing and the way claims play out.
Defense within limits (DWL, "wasting limits," "burning limits")
Defense costs are paid out of the policy limit. Every dollar spent on defense reduces the dollars available for indemnity. Standard structure for most professional liability lines. Cheaper than outside-limits defense because the limit is being consumed by both buckets.
Defense outside limits (DOL)
Defense costs are paid in addition to the policy limit. The full limit remains available for indemnity. More common in some specialty E&O lines (e.g., lawyers' professional in some states, miscellaneous professional). Significantly more expensive.
Why this matters operationally
In a long-tail claim that fights through motion practice, expert discovery, and trial, defense alone can run into the millions. On a $5M policy with defense within limits, an aggressive litigation can cut available indemnity in half before settlement is even on the table. On the same $5M policy with defense outside limits, the full $5M is still there for settlement. Insureds who litigate aggressively are very different consumers of the two structures, and underwriters price them accordingly.
Some forms are hybrids: defense within limits up to a point, then outside limits, or vice versa. Some forms put defense within limits but cap defense at a percentage of the limit. The wording governs. The marketing brochure does not.
Hammer clauses and consent
The "hammer clause" (also called the "consent to settle" clause) governs what happens when the carrier wants to settle a claim and the insured does not. Professional services firms care about this clause because their reputation is on the line and a settlement can read as an admission of error.
Three flavors
- Full hammer. The carrier can settle whenever it chooses, but if the insured refuses to consent, the insured pays the difference between the amount the carrier could have settled for and the eventual outcome (settlement or judgment), plus all defense costs from the date of refusal forward. This is the carrier-friendly version.
- Soft hammer. If the insured refuses to consent, the insured is responsible for a percentage (commonly 20-50%) of the additional loss and defense costs above what the carrier could have settled for. Compromise.
- No hammer / dual consent. Settlement requires both parties' consent. Dispute resolution mechanisms (binding arbitration, third-party evaluators) handle disagreements. The insured-friendly version, mostly seen in lawyers' and accountants' professional and in negotiated programs.
The economics of the hammer
From the carrier's perspective, the hammer prevents the insured from holding out for vindication at the carrier's expense. From the insured's perspective, a settlement can damage reputation and trigger collateral consequences (state bar reporting, license issues, NPDB reporting for healthcare). The hammer clause is one of the negotiated points on E&O placements above commodity premium thresholds.
The major sub-classes
"Professional liability" is an umbrella for a long list of sub-classes, each with its own form, its own appetite, and its own claims culture. The biggest:
Why the sub-classes matter for underwriting
Each sub-class has its own loss frequency curve, its own severity distribution, its own claim culture, and its own application. An accountant's policy and a lawyer's policy share a structural skeleton but the rating factors, the questions on the application, and the exclusions are entirely different. A miscellaneous PL form covering a consultant looks more like a manuscript wording each time, because the activities the consultant is performing have to be defined explicitly. Carriers that write across sub-classes maintain separate underwriting teams for each, with separate appetite memos.
Tech E&O and the cyber overlap
Tech E&O covers a software vendor, IT services provider, or technology consultancy when their work product fails their customer. Cyber covers the insured's own data environment when it is breached. For technology businesses these are adjacent risks, and the market has responded by bundling them.
The bundled tech / cyber form
Most modern tech E&O placements include cyber on the same policy. The structure: a single set of insuring agreements covering both professional services failures (the E&O part) and the insured's own privacy and security events (the cyber part). Policy limits can be combined or split by section. The advantage is no coverage gap and no allocation fight. The disadvantage is a single limit eroding to two different loss types.
Where the line blurs
A SaaS vendor ships a release that exposes customer data due to a coding error. Is that a tech E&O claim (defective work product), a cyber claim (privacy event), or both? Modern bundled forms answer "both, on the same policy, with one limit." Standalone forms have to allocate, and the allocation fights show up in claims. The trend is firmly toward bundled placement for tech businesses, but standalone placements still exist for the largest accounts where capacity dictates separate towers.
The application data for tech E&O looks very different from a standard professional liability application. Expect: revenue split by service vs subscription vs perpetual license, customer concentration, a list of named contracts with size and indemnity language, prior versions and product release notes, security control questionnaire, SOC 2 reports, sub-processor lists. Extracting and normalizing this is meaningfully harder than a standard PL submission, and is one of the highest-value automation targets in financial lines.
How underwriters evaluate the risk
The PL underwriter's job is to price the probability that an insured will make a costly mistake during the next year. The inputs are roughly the same across sub-classes:
- Revenue and revenue mix. Total revenue drives the rate base. Mix by service line drives the rate factor (audit work is priced differently from advisory; M&A work differently from real estate).
- Loss history. Five to ten years of claims and circumstances. Carriers care about both frequency and severity. A pattern of small frequency claims is a different risk from one large severity loss.
- Service area and client concentration. A single client concentration above 10-20% is a flag. A single matter concentration above 20% is a bigger flag.
- Quality controls. Engagement letter discipline, peer review, partner review thresholds, conflict-checking systems, supervision of junior staff, continuing education, malpractice prevention CLE.
- Specific risk areas. Sub-class specific. For lawyers: bankruptcy work, plaintiff securities, IP litigation, large transactional work. For accountants: SEC audits, broker-dealer audits, ERISA audits. For A&E: residential, hospitals, sports stadiums.
- Geography. Some venues are more plaintiff-friendly than others. Some states have caps. Some industries are concentrated in plaintiff-friendly venues.
The application and the supplementals
Standard PL applications run 8-15 pages. Sub-class supplementals add another 5-10 pages each. A large law firm submission can run 50+ pages with attachments (engagement letter samples, conflict-check policies, claim narrative for prior reported circumstances, partner CVs). Most of the data is unstructured narrative. The underwriter is trying to triangulate quality and discipline from documentary evidence rather than from physical inspection.
Pricing and capacity
Primary professional liability is rated on revenue with adjustments for service mix, claim history, and risk controls. Excess layers attach above the primary and follow form. Towers for large firms (especially law and accounting) can run $200M+ in aggregate, with capacity sourced from US, Bermuda, and London. Self-insured retentions on large accounts run from $1M to $25M+, depending on the firm's loss history and risk appetite.
Where IDP earns its keep
Professional liability submissions are document-heavy in a particular way: long applications, dense supplementals, prior loss narratives written in legalese, engagement letter samples, claim circumstance reports, contract attachments. The volume of unstructured text per submission is among the highest of any commercial line, and the structured-field count on the application is low. That is the IDP sweet spot.
The PL workflows where Indico shows up
- Retro date continuity check. Walk the prior policy declarations across multiple years and confirm the retro date is preserved at renewal. The single highest-leverage automation in PL renewal processing.
- Application warranty cross-check. Compare the no-knowledge warranty on the application against actual claim circumstances disclosed in the loss run. Flag inconsistencies for the underwriter before bind.
- Engagement letter analysis. Extract key clauses (indemnity, liability cap, governing law, hammer clause) from sample engagement letters and benchmark against industry norms.
- Loss run normalization. Different carriers issue loss runs in different formats (PDF, Excel, Word). Normalizing into a structured circumstance / claim history is the same problem across sub-classes.
- Circumstance narrative summarization. Most PL apps include a "describe any circumstance that could give rise to a claim" section. The narratives are dense, legalistic, and written by counsel. Summarizing into structured fields with source citations is genuine underwriter time savings.
The PL demo that opens doors is the retro date and warranty cross-check agent. The underwriter uploads three years of expiring declarations plus the new application, and the agent confirms (a) the retro date is preserved, (b) the warranty answers are consistent with the prior loss runs, (c) any new circumstances disclosed are properly carved out, and (d) the limit and retention structures are consistent. Five minutes of manual work compressed to seconds, and the audit trail back to the source document is what makes the underwriter trust it.