Written for Claims, operations and technology leaders, and the sponsors who have to fund a first-notice (FNOL) agent around a process that is already running.
What's in this playbook
This playbook is the operational document for a claims-intake (first notice of loss) agent. It is complementary to the Insurance AI Playbook already in this library, and is not a replacement for it. That playbook remains the industry map: six families, with FNOL as one of them, and a first case that puts two existing claims pilots on an inventory before a live claims use. This playbook will not reprint that map, and will not give a different reading of the same firm. The Agentic Workflows white paper remains the companion for a human gate as architecture; this playbook will not reprint its six patterns as deployments, and will not publish a migration roadmap.
What it is, instead, is a service playbook for the FNOL case: six families of first-notice work, fourteen design tests, and a first-case path consistent with that industry playbook. Intake, routing and acknowledgement are in-scope. Coverage, liability and settlement are a person’s decision. The human gate is architecture, not a prompt. A third claims pilot, an unsupervised settlement or deny, a coverage decision at intake, or a twelve-week claims framework, is not a first case this playbook will recommend. This playbook will not invent a “cycle time,” an “STP” rate, or an “accuracy” sector figure, and will not close a named insurance rule.
How the tests and the method are meant to be used
The financial method is The Complete Guide to AI Automation ROI. This playbook will not invent an FNOL-sector return, and it will not promise a twelve-week go-live. The 3.8x average first-year ROI across client engagements is an average, not a forecast or a commitment for any individual project. ROI is measured across client engagements against a baseline agreed before work starts. It is an average, not a forecast for any FNOL project, and is not used as a sector benchmark. Built to SOC 2 standards is the exact wording we use; it is not a certification of the reader’s system.
The illustrative firm is the same hypothetical mid-size property and casualty insurer as in the Insurance AI Playbook, the Enterprise AI Governance Framework, and the 2026 diagnostic: overall 15, Emerging, Governance the constraint at 1, with a next-step target of 24, Developing. Two claims pilots have been run; neither is in production with an operator. The honest first work is to put those two existing pilots on an inventory, with an owner, a sampling plan and a stop, and only then, if the tests pass, fund a bounded FNOL assist: extraction and completeness, with a person on anything that opens, pays, or denies. An unsupervised settlement or deny is not a first case. The target is a next level, not a timeline, and not a promise. A companion workbook, asymilink-claims-intake-fnol-playbook.xlsx, holds a family register, the fourteen tests, questions for counsel, a baseline sheet, and the same illustrative insurer. Do not wait to be Leading.
Six families of first-notice work
each with an honest first case, a case that is too wide, and a gate. Each item below is the shape of the map; the full wording, honest first cases and “too wide” warnings are in the PDF.
F1
Inventory of running claims systems (first)
F2
A bounded FNOL assist — extraction and completeness
F3
Routing and acknowledgement, with a person on the message
F4
A write into the claims system (later)
F5
An after-hours or volume assist that is still gated (later, and not the after-hours voice playbook)
F6
Hosting, decided before the case is funded
The PDF gives, for each family, an honest first case, a case that is too wide, and a defined gate. The summaries above are the shape of the map, not a substitute for that wording.
Fourteen design tests (excerpt)
Fourteen design tests are applied to every case. The answers will differ. The standard does not. The full wording is in the PDF and the companion workbook.
T1–early
a named inventory; a person on anything that opens, pays, denies, or sounds like coverage; counsel asked without a date taken from this document.
Mid-tests
a refusal to recommend an unsupervised settlement or a coverage decision at intake as a first case; a trail and a stop; hosting decided before the case is funded.
Later tests
a signed baseline on the ROI method; a write into the claims system as a later design, not a first case.
A completed test sheet is not a certificate, not a compliance score, and not a finding that a system is safe or aligned with a named regime.
Get the guide and workbook
Enter your name, work email and company. The files appear on this page immediately. A backup download link is also emailed when email delivery is available.