Lines of business
Rate commercial property by location, not by account
A commercial property policy is a schedule of places. Construction, protection class and territory belong to the location rather than to the business, so one policy can carry twenty locations that resolve twenty different ways and still has to produce one number you can defend. ISO commercial property rates on SelectRate today. Governed bureau content sits underneath, your carrier and program configuration sits on top, and a deviation is set once rather than twenty times across a schedule.
The unit of rating is the location
General liability usually describes one business doing one thing: a class, an exposure basis, a territory, a set of limits. Commercial property describes a list of addresses.
Each address carries its own construction, its own protection class and its own territory. Each one produces its own rate. The policy premium is what those location results add up to.
That single difference decides how the rating system underneath has to be built. A rater that treats a submission as one flat set of facts will do general liability well and property badly. Property needs the schedule treated as a structure rather than a flat form: each location resolving its own characteristics, and coverage written at the level where the value actually sits. That is the first thing to test when you evaluate a rater for property, ours included.
What has to resolve, location by location
| Characteristic | Where it belongs | Why it complicates rating |
|---|---|---|
| Construction | The building | Two buildings at one address can rate differently, so they cannot be averaged into a single location-level assumption. |
| Protection class | The location | It is an underwriting characteristic of the place that feeds the commercial property rate. It has to be carried per location and applied per location, never set once for the account. |
| Territory | The location | It is a filed mapping question asked once per location. Two locations in the same state can sit in different territories. |
| Values by coverage | The building and the coverage | Building, contents and business income are separate amounts, and premium has to trace back to the amount that produced it. |
Construction and protection class are facts about a place
An account does not have a construction type. A building does. An account does not have a protection class. A location does. The moment a system stores either of them at account level to make the data model tidier, the rate stops being a property rate and starts being an approximation that an underwriter has to correct by hand.
This is also why property schedules grow awkwardly. Add a location and you have not added a field, you have added another complete set of rating characteristics that has to resolve on its own and then join the total.
Territory is asked once per location
On general liability, SelectRate derives territory from the filed mapping rather than accepting it on the request, and reports the selection in the worksheet provenance. You do not send a territory. Commercial property resolves construction, protection class and territory by location, so the same question is asked once per location. The provenance record therefore has to stay readable per location rather than collapsing into one account-level answer.
Coverage written on values, not on a payroll or sales basis
General liability rates against a premium basis such as gross sales. Commercial property rates against values. Building, contents and business income are amounts the insured carries, and they move for reasons that have nothing to do with how busy the business is.
That changes both ends of the transaction. The input is a set of amounts by coverage by location, not a single exposure figure. The output has to show which amount produced which part of the premium, because that is the first thing an underwriter checks when a number looks wrong.
One carrier deviation, every location on the schedule
Layer two of the model is carrier configuration: carrier-approved deviations from the rating base, including company multipliers, set by state. Layer three is program configuration, the pricing and rules specific to your program. You set one multiplier for the whole program and override the single state that differs.
Property puts that mechanism under a particular kind of pressure. A single-state general liability risk answers the state override question once. A property account with locations in four states answers it four times, and all four answers have to come from the one program version you released. If they can drift apart, the account total is not reproducible, and no amount of worksheet detail repairs that afterwards.
Consistency across a schedule is not a reporting feature. It is the reason the configuration model is layered in the first place: your deviations sit in one place, above the bureau base, and the engine applies them rather than a person applying them twenty times.
What a multi-location worksheet has to show
SelectRate captures every rating step as the engine executes it, with the governing value and where that value came from. Provenance records which filed mapping was selected, why it governed, and the document and page it came from. Every intermediate value is shown at the precision the engine used.
A single-location general liability worksheet is a column of steps. A twelve-location property worksheet is that column repeated twelve times, plus the arithmetic that joins the columns together. Three things have to be legible that a monoline general liability worksheet never has to show. Use them as the checklist when you review a multi-location worksheet in a demo:
- Which location a step belongs to.
- Which characteristic of that location governed the step.
- How the location results became the policy result.
If you want the step-by-step reading of a worksheet, the general liability rating page walks one through in detail. Property is the same discipline applied to a schedule instead of a single risk.
Versions, effective dates and the layer underneath
Property programs live for years, and the content underneath them keeps moving. When a circular issues, the base layer moves underneath you and your configuration stays intact. Creating a change never disturbs what is live. A version is published with an effective date, and the change is live in Workbench and the Rating API at once.
Property gets more out of that than most lines, because schedules move. Locations are added between quote and bind, values are corrected, a building comes off. Prior versions stay callable for audit, and a quote from three years ago re-rates on the version and edition it was written under, so the number you defend later is the number you gave.
A rating run resolves the version in force for the policy’s effective date. There is no version to pin and no way to rate on one that was not in force.
Test the schedule before you release the program
Every change is tested in Program Studio before it can go forward: rate complete test risks against the draft version and against what is published, and read the worksheet behind each result. A version cannot be submitted for release until at least one test has rated, and the review step shows exactly what it changes against the published configuration, state by state. A published version takes effect on its effective date and never changes afterwards. A program that is not effective for the quote date returns no premium, rather than quietly answering from the wrong version.
Test with a real schedule. A property program that behaves on a clean single-location sample has told you almost nothing about how it will behave on an account with nine locations, three states and a building that should have come off last year.
One endpoint, granted by program
The Rating API is synchronous JSON running the same engine as Workbench: identical inputs, program and effective date produce an identical premium and worksheet. Each API client has its own credential and is granted the specific programs it may rate. Scopes decide which operations a credential can call. Grants decide which programs. There is no wildcard at request time, and a client never sees another client's quotes or policies. Every rating channel you authorize, your own underwriters included, works off the version you released.
Products are discovered rather than hard-coded. GET /products/{productCode}/schema returns the field metadata for a product's risk payload: names, types, options, which fields are required and conditional visibility. New lines of business appear through discovery without a contract change, so the shape of a property risk payload reaches your integration as data rather than as a renegotiation. Full request and response detail is in the Rating API documentation.
Start from the standard
A property program is the wrong place to begin with an empty configuration screen. Every organization on SelectRate has a Standard Commercial Property program: a governed program sitting directly on the published ISO content, ready to inspect before you commit to anything.
It answers the question a launch plan actually turns on, state by state. Which states are in force. Which are ready to rate now. Which are waiting on a company value, and precisely which value, whether that is a loss cost multiplier, a company factor or the adoption date of a company dated edition. Nothing is guessed on your behalf and nothing is hidden.
You run Program Test on it, rate the sample transactions it carries, and when your values are agreed you create your own program from it. Configuring what makes your program different is the work. Rebuilding the bureau foundation is not.
Governed foundation
Published ISO commercial property content, effective dated and maintained by Selectsys. Read only.
Standard Commercial Property
One per organization. States, readiness, the exact company value each state still needs, and Program Test.
Your carrier program
Created from the standard, carrying your multipliers, adoption dates and authorized deviations.
The Workbench adapts to the rating product
Underwriter Workbench is not a hand built property screen. The intake comes from the rating product itself, which is why a property submission looks like a property submission: locations that repeat, buildings that repeat inside a location, and coverage amounts entered where the value actually sits.
Governed classifications and option lists are resolved for the state and the effective date rather than typed from a reference sheet. Where the content sends a risk to an underwriter, it comes back as a referral with its reason rather than as a failed request. The same product shape reaches your own systems through the Rating API, so nothing about property is a one off.
This is the same Workbench that carries general liability, businessowners, commercial auto and the AAIS products. One screen adapts, rather than each line getting its own.
Rate through the policy term
Property schedules move after bind more than most lines. A location is added, values are corrected, a building comes off, a term is cancelled and reinstated. Rating has to hold up through all of it.
SelectRate rates new business, renewal, endorsement, cancellation, reinstatement and rewrite on commercial property. An endorsement rates on the versions the term was written on, so a version published later never changes an in force term, and the premium effect is the difference across the whole term. A renewal rates on the versions in force at the renewal date. A cancellation is previewed before it is confirmed, with the return premium and the earned premium shown per component.
Every accepted transaction stays in the term ledger with its own calculation and worksheet. How lifecycle rating works
The rest of the book
Commercial property runs on the same configuration model as general liability, businessowners, commercial auto and the ten AAIS products: governed bureau content underneath, your carrier and program configuration on top. An account that needs property alongside the AAIS Commercial Output Program is written in the same place, in the same Workbench, and searched in the same policy search.
Questions buyers actually ask
Does an account with locations in several states have to be split into several quotes?
No. One quote carries the whole schedule. Each location resolves its own construction, protection class and territory for its own state, and the location results join into one account premium with one worksheet behind it.
Do we send a territory for each location?
No. Territory is derived from the filed mapping per location and reported in the worksheet provenance, including on a schedule whose locations sit in different territories in the same state.
How do we check a property number three years from now?
Rate a risk as of any date. Prior versions stay callable for audit, a quote re-rates on the version and edition it was written under, and the worksheet holds the governing value behind every step and where that value came from.
What should we bring to a demo?
Bring a schedule with real locations in more than one state, ideally a messy one. We will rate it on the call and read the worksheet with you, location by location.
Bring a schedule, not a slide
The fastest way to judge a property rater is to put a messy account through it and read the worksheet. Bring a schedule with locations in more than one state and we will rate it with you on the call.