← Back to the Lab
[ 04 ]Deployment Ready · Service & Retention · Renewal

Retention Radar

Combines premium movement, service-ticket sentiment, and engagement signals into a single churn score, then routes the riskiest books to the right producer with a suggested play.

PredictiveRetentionAnalytics

[ capabilities ]

  • —90-day churn probability per account
  • —Automatic producer task assignment
  • —Book-level retention forecasting

[ core systems & integration architecture ]

Scores the book from policy, billing, and service history, then drops a prioritized task on the right producer inside the system they already live in.

Applied Epic

Native REST API

Policy, activity, and premium history reads; task write-back

Vertafore AMS360

Native REST API

Book and suspense data via Orange Partner APIs

Guidewire BillingCenter

Event Stream / Webhook

Late-payment and cancellation events as churn signals

Salesforce FSC

Embedded Sidecar

Churn score surfaced on the account record

Sync model

Nightly scoring refresh with real-time event overrides

Standards spoken

JSON RESTWebhooksSQL warehouse export

Time to connect

Days

Enterprise guardrails

  • —No customer data used to train models — zero-retention inference only
  • —OAuth 2.0 / mTLS against your existing identity provider
  • —Field-level audit trail on every value written back

[ mcp microservice interface ]

This module ships as an MCP microservice — retention-mcp. It consumes governed tools from the core-system adapters and publishes its own, so your assistant, orchestrator, or another module can call it directly. No credentials ever reach the model runtime, and every write lands in the human gate.

Consumes from core systems

  • epic_get_policy_schedule
  • epic_get_activity_history
  • gw_get_claim_timeline

Provides to other agents

  • score_account_riskread

    Churn likelihood with the signals that drove it

  • list_at_risk_bookread

    Rank a producer's book by retention risk

  • stage_outreach_taskwrite · human gate

    Create a save-the-account task for the named producer

See how core APIs become MCP microservices →

[ implementation playbook ]

How an agency actually stands Retention Radar up — read-first, human-gated, and measured from day one.

Setup, in order

  1. 1

    Connect read access to your book of business — policies, renewal dates, premiums, and claims history.

  2. 2

    Define your risk signals: premium increase thresholds, claims in the last 12 months, coverage reductions, and how long since anyone touched the account.

  3. 3

    Set the radar window — most agencies start at 120 days before renewal.

  4. 4

    Route alerts to the account owner, not a shared inbox. Retention dies in shared inboxes.

  5. 5

    Run the first month in review mode: the radar flags accounts, you check whether the flags were right, and tune the thresholds.

  6. 6

    Add the outreach step: each flag comes with a drafted talking-points brief so the producer calls with context, not a cold check-in.

Agency use cases

The 90-day at-risk list

Every Monday, each producer gets their accounts renewing in 90 days that show risk signals — rate shock, a recent claim, or radio silence — ranked by premium at stake.

Win-back after a lost account

When an account leaves, the radar tracks the x-date and queues a win-back brief eleven months later, while the new broker is complacent.

Service-team early warning

A spike in service tickets or a complaint email flags the account before the renewal conversation, so the producer walks in knowing there's repair work to do.

The first win to aim for

Save one account in the first quarter that would otherwise have shopped — that's usually the whole year's ROI.

Guidance by agency role

01

Producers

Work the highest-premium risk alerts first, use the prepared talking points to make a personal call, and record whether the signal led to an intervention.

02

Account managers

Validate service and claims signals, add account history the radar cannot infer, and coordinate the recovery plan before the producer reaches out.

03

Agency administrators

Set risk thresholds and ownership rules, audit untouched alerts, and compare flagged-account retention with the rest of the book each month.

[ built for ]

01Agencies
02Brokers
03Carriers

[ where this sits in the vendor market ]

We recommend buying first wherever the market already solves a problem well. Here is what you can buy instead of building this — and where this module earns its place beside what you buy.

No product in the directory covers this yet

That is why Retention Radar exists — the market has not packaged this work, so the module is the practical way to get it.

Browse the independent vendor directory →
What would Retention Radar hand back to your team?Run it through the lab ROI calculator — hours, dollars, and the cost of waiting.