Algorithmic broker

Category: Tech distribution · Reviewed by Jake Leat, Associate Director · Last reviewed 2026-06-10

An algorithmic broker is an FCA-authorised insurance intermediary that uses automated rules, decision trees or machine-learning models to place risks with one or more insurers, typically without a traditional manual underwriting submission and often within seconds of receiving the customer’s information.

Algorithmic placing has grown rapidly in UK small and medium-sized enterprise (SME) commercial insurance, with platforms such as Send Technology, hyperexponential, Cytora, Artificial Labs and Concirrus connecting brokers to insurer pricing models via APIs. The FCA Handbook’s Principles for Businesses (in particular PRIN 2.1.1R) and the Consumer Duty (PRIN 2A) impose the standards against which automated placement is judged, alongside ICOBS 5 (demands and needs) and PROD 4 (product governance).

Definition

An algorithmic broker is an intermediary whose business model involves:

• Capturing the customer’s risk information through digital channels (web form, broker platform or API).
• Applying rule-based or model-based logic to identify suitable insurer(s) and pricing.
• Generating a quotation or binding decision without traditional underwriting submission.
• Maintaining model governance, audit trails and human oversight under SYSC.

It is distinguished from a price comparison website by its role on the broker side of the transaction (placing the risk with insurers) rather than the consumer-facing comparison role.

Legal and regulatory basis

The principal UK framework comprises:

• Financial Services and Markets Act 2000 — perimeter; the activity is typically “arranging deals in” non-investment insurance contracts (see FSMA 2000).
• FCA Handbook PRIN, PRIN 2A — the Principles for Businesses including the Consumer Duty for retail business.
• FCA Handbook ICOBS 5 — demands-and-needs identification.
• FCA Handbook PROD 4 — manufacturer / distributor product governance, including fair value assessments.
• FCA Handbook SYSC — systems and controls, model risk management.
• FCA PS17/27 “General Insurance Distribution Chain” — ensuring fair value across distribution chains.
• EIOPA AI Governance Principles (2021).
• FCA AI Update (April 2024).

How it works in practice

A typical algorithmic placement journey:

1. Risk capture — broker platform or direct API receives policyholder data.
2. Augmentation — external data (Companies House, Experian, geospatial, claims databases) appended.
3. Eligibility filters — rule-based exclusion of out-of-appetite risks.
4. Quotation engine — insurer pricing models, often via API, return quotes within seconds.
5. Bind decision — algorithm selects the optimal placement (price, terms, broker remuneration) and binds, or surfaces to a human for review.
6. Documentation — IPID, policy summary, demands-and-needs statement and PRIN 2A consumer outcome evidence retained.

For commercial business outside the Consumer Duty perimeter, the PRIN baseline and ICOBS still apply.

Common variations and subsequent developments

• SME e-trading platforms — Acturis, Open GI, Applied Epic broker-managed quote and bind, increasingly with embedded algorithmic placement.
• Lloyd’s Blueprint Two — modernised market infrastructure with API-based placement and digital binders.
• Algorithmic placement reinsurance — facultative reinsurance via algorithm-based platforms (e.g. Akinova, Ki).
• Ki Syndicate — Ki, a Lloyd’s algorithmic follow syndicate launched 2021, places risks via API based on a lead pricing signal.
• Model risk management — SYSC expectations following the Bank of England SS1/23 on model risk principles (banking, with read-across for insurers).

Example

A UK SME insurance broker integrates a placing platform with five participating insurers. A small manufacturer’s property policy is quoted by capturing the risk in 12 fields, augmenting with Ordnance Survey and Companies House data, and calling the insurers’ API pricing engines. Three quotations return within 15 seconds; the algorithm selects the best fair-value option taking into account price, cover, claims service and broker remuneration consistent with PROD 4. The bind is recorded, the IPID and policy summary delivered, and PRIN 2A consumer outcome metrics collected. A human broker reviews any case where eligibility filters surface anomalies.

See also

• API marketplace insurance
• Robo-advisory insurance
• Aggregator insurance
• Price comparison website
• Consumer Duty
• ICOBS
• Principles for Businesses
• Lloyd’s
• Managing general agent

References

• FCA Handbook: PRIN, PRIN 2A, ICOBS 5, PROD 4, SYSC.
• FCA, PS17/27 “General Insurance Distribution Chain”, September 2017.
• FCA, “Artificial Intelligence Update”, April 2024.
• Bank of England, SS1/23 “Model risk management principles for banks”, May 2023.
• EIOPA, “Artificial Intelligence Governance Principles”, June 2021.
• Lloyd’s, “Blueprint Two: the future at Lloyd’s” (2020, updated 2024).
• Financial Services and Markets Act 2000, legislation.gov.uk.
• Insurance Distribution Directive (Directive (EU) 2016/97), as retained.

This entry is part of the Apex Insurance Wiki. Last reviewed by Matt Bartlett on 2026-06-10. Next review: 2026-12-10.

Apex Insurance Brokers Limited. Authorised and regulated by the Financial Conduct Authority, FRN 724952. Registered in England and Wales, Companies House 07014570. This entry provides general information about UK insurance concepts and is not regulated advice. Consult your insurance broker on your specific position.