# Rating & Quoting

Take a cleared broker submission from package to a drafted quote — rated, checked against authority, and ready for an underwriter's decision.

## The business problem

Once a submission clears triage, the underwriter's work is mostly assembly before it's judgment: rating every line, stratifying losses into a pick, checking the result against authority and treaty capacity, and drafting the quote letter with its subjectivities. That assembly is mechanical, and it's where most of the time goes on a file the underwriter will ultimately just decide on.

The **Rating & Quoting** template automates the assembly: it reconciles the account, rates every requested line, stratifies the loss history, checks authority, and drafts a two-option quote — leaving the underwriter to review and decide rather than build.

> **Important: the quote is decision support, not a decision.** The agent prepares the pricing and the reasoning behind it; it doesn't release a quote or bind coverage. An underwriter reviews, edits, and approves every item before it's finalized.

## Trigger and source data

The template ships with the **Manual File Upload** trigger. You can configure other triggers, including a connection to an Outlook email inbox. Inbound webhooks, scheduled/timer triggers, and SFTP are **not currently available**; drive runs from another system through the [Bevaya API](../api/flow-executions#run-a-flow). See [Configure triggers](../build-ai-agents/triggers.md).

Upload a cleared broker submission package — typically the same package used for [Triage and Routing](./triage-and-routing.md) (application, coverage-line supplements, schedules, loss runs, SOV), plus:

- The **expiring policy or declarations**, to anchor pricing
- **Loss control or inspection reports**, if available, for risk control subjectivities

## Template starting point

Create an AI agent from the **Rating & Quoting** template for a working flow that extracts, rates, and drafts a quote out of the box. Open it in the canvas, select an environment, click **Start Editing**, and tune it to your book — the rate book, authority tiers, and forms. See [Create an AI Agent](../build-ai-agents/create-agent.md).

## What the template preconfigures

| Step in the flow | Node it uses | Category |
| --- | --- | --- |
| Prepare the uploaded files | [Read Files](../build-ai-agents/utility-nodes.md) | Utility |
| Process each document in the package | [For Loop](../build-ai-agents/control-nodes.md) | Control |
| Classify each document and extract per-type fields | [InsurGPT: Custom](../build-ai-agents/insurgpt-nodes.md) | InsurGPT |
| Route each document to the right extractor | [Switch](../build-ai-agents/control-nodes.md) | Control |
| Reconcile the account and the loss history for pricing | [InsurGPT: Custom](../build-ai-agents/insurgpt-nodes.md) | InsurGPT |
| Check extracted values against confidence rules | [Field Validation](../build-ai-agents/utility-nodes.md) | Utility |
| Rate every line, stratify losses, check authority, and draft the quote | [Custom Code Block](../build-ai-agents/utility-nodes.md) | Utility |
| Route referrals and standard quotes to separate queues | [Switch](../build-ai-agents/control-nodes.md) | Control |
| Route the prepared quote to an underwriter | [Review](../build-ai-agents/utility-nodes.md) | Utility |
| Close out the item | [Complete](../build-ai-agents/action-nodes.md) | Action |

The rating step produces the findings shown to the underwriter — a status (Quote Ready, Referral Required, or Not Quotable), two quote options with premium by line, the loss pick, the authority tier required, and a drafted quote letter with subjectivities.

## What builders must configure before deployment

- **Rating.** The template ships with a Custom Code Block pricing off a rate book held as data, so you can tune it to your book without touching the flow logic. For a complex or frequently changing rate plan, you can instead have an InsurGPT node price from your rate manual directly. Either way, replace the demo rates before relying on the output.
- **A connection to your formal rater, if you want one.** We recommend an [HTTP node](../build-ai-agents/utility-nodes.md) to send reconciled exposures to your rating engine and bring the rated premium back into the flow, keeping pricing on your filed, governed logic.
- **Authority and treaty limits.** Set your letter of authority tiers and treaty capacity by line.
- **Forms, endorsements, and subjectivities.** Map your policy forms and standard subjectivities by line.
- **Review routing.** Assign reviewers for the standard quote queue and the referral queue. See [Utility nodes](../build-ai-agents/utility-nodes.md).

## How the AI agent behaves

The agent classifies and extracts each document, reconciles the account (including prior years of premium, losses, and claim counts), and rates every requested line — never assuming an exposure or loss year is fine just because a document exists. It stratifies losses into an annual pick, checks the result against authority and treaty capacity, and drafts a quote letter with subjectivities. It works only from the supplied documents (and your rater, if connected); it doesn't invent values, and low-confidence or missing fields are flagged rather than fabricated.

## Human review model

The underwriter works the item in the Bevaya Platform across two tabs:

- The **Insights tab** — both quote options, the authority required, price adequacy and referral reasons as key insights, and an underwriting summary.
- The **Review tab** — field-by-field verification of the account, the pricing, and the loss history, each linked to its source document and page. The underwriter can edit the quote letter and complete the referral note before approving.

Every quote pauses at **Review**; referrals land in a separate queue from standard quotes. Reviewers can work items and reviews but cannot build or run flows. See [Human review](../monitor-review/human-review.md).

## Run status and reporting

Each submission becomes an item moving through **In Progress, Review, Complete, Failed, Canceled**. Watch runs in [Run history](../monitor-review/run-history.md) and live counts by status in [Item status reporting](../monitor-review/item-status-reporting.md).

## Example run

A cleared submission for a coastal resort arrives with an application, a statement of values, a loss control survey, the expiring declarations, and five years of loss runs.

1. The flow classifies and extracts each document, then reconciles the account and the loss history.
2. Rating produces two quote options; the loss pick drives the quote below manual premium.
3. The exposure and pricing require senior authority, so the item routes to the referral queue with a drafted referral note.
4. The underwriter reviews the loss pick and the survey findings, adjusts the quote, and approves.

## Common failure modes

- **Not Quotable from missing data.** A misclassified document or a miscounted loss run year stops the quote — check the extracted fields first.
- **Thin loss history.** When loss runs don't break out policy years, the loss pick leans on manual premium and the summary says so.
- **Rate book out of calibration.** An uncalibrated rate book prices an account away from market, producing referrals that reflect the rates, not the risk.
- **Low-confidence or missing fields.** Flagged by validation rather than fabricated.

## Recommended rollout path

1. **Replace the demo rate book**, or connect your formal rater, before relying on the output.
2. **Set authority tiers and treaty capacity** to match your book.
3. **Build and test** on sample submissions with **Run Draft**, inspecting each step. See [Test and debug a draft](../build-ai-agents/test-debug-draft.md).
4. **Validate pricing** against accounts your underwriters have already quoted.
5. **Pilot with every quote reviewed** to build trust before adjusting review routing.
6. **Promote to production.** See [Drafts and publishing](../build-ai-agents/drafts-publishing.md).

## Where to go next

- [Triage and Routing](./triage-and-routing.md) — the step before, which decides whether a submission reaches rating.
- [Policy Renewal Handling](./policy-renewal.md) — the equivalent pass for an expiring policy instead of new business.
- [Utility nodes](../build-ai-agents/utility-nodes.md) — Field Validation, Custom Code Block, and Review.
- [InsurGPT nodes](../build-ai-agents/insurgpt-nodes.md) — classification, extraction, and reconciliation.
- [Human review](../monitor-review/human-review.md) — the Insights and Review tabs an underwriter works from.
- [Item status reporting](../monitor-review/item-status-reporting.md) — track submissions by status.
- [Use cases overview](./overview.md) — the full catalog of supported patterns.