Book Intelligence Warehouse
Consolidates policy, commission, and activity data into one warehouse with a semantic model, so leadership can answer growth and profitability questions without a spreadsheet marathon.
[ capabilities ]
- —AMS, carrier, and commission ingestion
- —Semantic model for self-serve reporting
- —Historical snapshots for trend analysis
[ core systems & integration architecture ]
The data foundation the other modules read from. Consolidates policy, commission, and activity data without touching production transaction processing.
Guidewire Data Platform
Batch & File DropCloud Data Access extracts into your warehouse
Duck Creek OnDemand
Batch & File DropScheduled extract with canonical schema mapping
Applied Epic / AMS360
Native REST APIIncremental policy, commission, and activity pulls
Carrier commission statements
ACORD / IVANS BridgeStatement parsing reconciled against policy records
Sync model
Nightly incremental load with historical snapshots
Standards spoken
Time to connect
Weeks
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
- —Read-only extracts — production systems are never written to
[ mcp microservice interface ]
This module ships as an MCP microservice — warehouse-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
- ivans_latest_download
- gw_get_policy_period
Provides to other agents
query_bookreadStructured analytics query across the whole book
segment_accountsreadSegment the book by class, carrier, premium or loss ratio
[ implementation playbook ]
How an agency actually stands Book Intelligence Warehouse up — read-first, human-gated, and measured from day one.
Setup, in order
- 1
Connect your AMS as the primary source — the warehouse reads, it never writes back.
- 2
Add the secondary sources: carrier downloads, commission statements, and the spreadsheets that secretly run your agency.
- 3
Let the warehouse normalize everything into one model: clients, policies, premiums, commissions, claims, producers.
- 4
Validate the numbers against a month-end report you trust before anyone makes a decision from the warehouse.
- 5
Build the first five dashboards with the leadership team: book growth, retention, carrier mix, producer performance, revenue per employee.
- 6
Open self-serve querying so anyone can ask questions of the book in plain language.
Agency use cases
The Monday leadership dashboard
Book growth, retention rate, and pipeline — current to last night, not last month's export — land in one view the leadership team actually checks.
Carrier concentration answers in seconds
When a carrier asks about your mix, or a market event hits, you know exactly how much premium sits with each carrier and in each line — no spreadsheet archaeology.
Acquisition due diligence
When evaluating a book to acquire, the warehouse ingests the seller's data and answers the real questions — retention, commission quality, account concentration — in days instead of weeks.
The first win to aim for
Kill the month-end reporting scramble: the numbers leadership asks for are always current and always reconciled.
Guidance by agency role
Producers
Use book and pipeline views to prioritize growth and retention, and correct ownership or stage data at the source when a dashboard exposes a gap.
Account managers
Use renewal, concentration, and coverage-gap views to plan service work, while flagging source records that do not match client reality.
Agency administrators
Own definitions, refresh checks, and monthly reconciliation; publish a data dictionary so every team reads retention, premium, and revenue consistently.
[ built for ]
[ 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.
Products that already cover this ground
Benchmarking and data services for agencies
Buy it when: Owners who want to know how their book compares to peers and where to grow.
Where it falls short: Teams needing their own queryable warehouse of their own data.
Data warehouse and analytics built on the Applied data model
Buy it when: Applied shops that want reporting without standing up their own warehouse.
Where it falls short: Agencies whose systems span multiple vendors and need a neutral model.
Works alongside products you already own