# Policy Renewal Handling

Read a renewal file, compare this year's exposures against last year's, price the renewal, and decide: renew, renew on conditions, or recommend non-renewal — with the statutory deadline already worked out.

## The business problem

Every expiring policy needs the same re-underwriting pass: compare exposures, check the premium audit, read current-term losses, confirm loss control recommendations were completed, and decide a rate change — all before the broker starts remarketing, and against a legal clock, since most states require advance notice to non-renew or to renew on materially worse terms.

The **Policy Renewal Handling** template automates that pass: it reconciles expiring against renewal exposures, checks the statutory clock, prices the renewal from expiring premium, and segments the account into Auto-Renew, Low-Touch, or High-Touch. Auto-Renew accounts complete without review; everything else reaches an underwriter with the analysis, the deadlines, and a drafted renewal letter or notice.

> **Important: the renewal is decision support, not a decision — with a narrow straight-through lane.** The agent recommends a decision and drafts the notice; it doesn't issue anything. Only clean, confident accounts complete without review, and reviewers can override any value.

## 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 renewal file — an email with attachments, a zip, or a combined PDF — typically:

- The **expiring policy declarations**
- **Internal loss runs** with adjuster notes
- **Loss control recommendation status**
- The **premium audit** (estimated versus audited exposures)
- **Endorsement history**
- The **renewal application and schedules**

## Template starting point

Create an AI agent from the **Policy Renewal Handling** template for a working flow that reads a renewal file and prepares the decision out of the box. Open it in the canvas, select an environment, click **Start Editing**, and tune it to your book — the statutory rules, rate guidance, and segmentation thresholds. 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 file | [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, loss years, and expiring versus renewal figures | [InsurGPT: Custom](../build-ai-agents/insurgpt-nodes.md) | InsurGPT |
| Check extracted values against confidence rules | [Field Validation](../build-ai-agents/utility-nodes.md) | Utility |
| Check the statutory clock, reconcile exposures, price, and segment | [Custom Code Block](../build-ai-agents/utility-nodes.md) | Utility |
| Route Auto-Renew straight through, others to the right queue | [Switch](../build-ai-agents/control-nodes.md) | Control |
| Route the prepared renewal to an underwriter | [Review](../build-ai-agents/utility-nodes.md) | Utility |
| Close out the item | [Complete](../build-ai-agents/action-nodes.md) | Action |

The renewal step produces the findings shown to the underwriter — a decision (Renew, Conditional Renewal, Non-Renewal Recommended, Renew Required By Statute, or Pending Renewal Information), a track (Auto-Renew, Low-Touch, or High-Touch), the notice deadline, exposure changes by line, current-term losses and recommendation compliance, two renewal options, and a drafted renewal letter or notice.

## What builders must configure before deployment

- **Statutory rules.** The template ships with an illustrative notice table. Replace it with rules your compliance team maintains, by state, line, and notice type, before relying on any deadline.
- **Rate guidance.** Set your performance tiers, rate bands, and maximum rate change — or connect your pricing engine with an [HTTP node](../build-ai-agents/utility-nodes.md).
- **Segmentation thresholds.** Set the Auto-Renew criteria (premium, loss ratio, exposure change, rate range, days to expiration) and the High-Touch triggers.
- **Appetite screen.** Reuses the triage appetite rules — replace the demo appetite guide there too.
- **Review routing.** Assign reviewers for the standard and High-Touch queues, and decide whether Auto-Renew completes without review. See [Utility nodes](../build-ai-agents/utility-nodes.md).

## How the AI agent behaves

The agent classifies and extracts each document, then reconciles the account, the loss history, and the renewal picture — separating what the expiring policy was written on, what the broker reports now, and what the audit found. It measures days to expiration against statutory notice windows, checks exposure change and loss control compliance, prices the renewal from the exposure-adjusted expiring premium, and segments the account. It works only from the supplied documents; missing values are flagged, not filled.

## Human review model

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

- The **Insights tab** — the recommended action, renewal premium and premium change, the statutory clock, rate change, and loss control compliance as key insights, and a renewal summary.
- The **Review tab** — field-by-field verification of the decision, pricing, exposure review, and loss and compliance findings, each linked to its source document and page. The underwriter can edit the drafted renewal letter or notice before approving.

High-Touch accounts land in a separate referral queue from Low-Touch renewals. 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 renewal file 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

An oceanfront resort's renewal file arrives 105 days before expiration with the expiring declarations, internal loss runs, a recommendation status report, and the broker's renewal application.

1. The flow reconciles the account and finds an open named storm claim and a critical roof recommendation past its due date.
2. The appetite screen passes; insurance to value comes out low, so the flow recommends raising building values.
3. The account prices with a compliance debit, and the rate change exceeds the state's conditional-renewal threshold.
4. The flow recommends a **Conditional Renewal**, states the notice deadline, and routes the item to the referral queue with a drafted renewal letter.

## Common failure modes

- **Deadlines from the illustrative table.** The statutory rules ship as a demo — replace them before relying on any deadline.
- **Expiring premium not split by line.** If declarations show only a total, the flow prices at account level and can't measure exposure change by line.
- **First renewal with no completed audit.** Without a final audit, the exposure base relies on the broker's estimate.
- **Recommendation status inferred.** A status log without due dates weakens the compliance check.

## Recommended rollout path

1. **Replace the statutory table** with compliance-maintained rules, and have compliance sign off on the deadline logic.
2. **Calibrate rate guidance** against renewals your underwriters have already priced.
3. **Build and test** on sample renewal files with **Run Draft**, one per expected track. See [Test and debug a draft](../build-ai-agents/test-debug-draft.md).
4. **Pilot with Auto-Renew turned off**, so a person reviews every renewal while you build trust in the segmentation.
5. **Enable Auto-Renew** for the small, clean segment once the pilot shows it's reliable.
6. **Promote to production.** See [Drafts and publishing](../build-ai-agents/drafts-publishing.md).

## Where to go next

- [Rating & Quoting](./rating-and-quoting.md) — the equivalent pass for new business.
- [Policy-to-Policy Comparison](./policy-to-policy-comparison.md) — checking the renewal against the expiring policy, attribute by attribute.
- [Triage and Routing](./triage-and-routing.md) — screening new submissions before they reach underwriting.
- [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 renewals by status.
- [Use cases overview](./overview.md) — the full catalog of supported patterns.