Reference
What is an ISO rater?
An ISO rater is software that calculates commercial property and casualty insurance premiums using licensed bureau content from Insurance Services Office. It is used by carriers, MGAs and wholesalers to price commercial risks consistently and to show how each premium was arrived at.
First, the two things called an ISO rating
The phrase gets used for two unrelated things, and search results mix them freely, so it is worth separating them before anything else.
Commercial insurance rating. ISO publishes advisory loss costs, rules and classifications for commercial property and casualty lines. Insurers license that content and build their own pricing on top of it. Software that performs those calculations is an ISO rater. That is the subject of this page and of this site.
Public Protection Classification. ISO also grades the fire protection capability of communities, which is why a property owner might be told their address has an "ISO rating" of some number. That grading feeds into property underwriting, but it is a community score, not a piece of rating software, and it is not what an ISO rater is.
If you arrived here looking for a protection class or a fire department grading for an address, this is not that. The rest of this page is about the software.
What an ISO rater actually does
An ISO rater sits between a description of a risk and a premium somebody can bind.
It takes the facts about a risk: what the business does, where it operates, what it owns or drives, what limits it is buying. It resolves the parts of the rating that are not supplied but derived, such as which territory a location falls in. It applies the licensed bureau content that is in force for the policy's effective date. Then it applies the insurer's own pricing on top, because bureau content is advisory and no carrier prices at the bureau level without adjustment.
What comes out is a premium and, in a system worth using, a record of how the premium was reached.
That last part is the difference between a calculator and a rater. Anyone can multiply numbers together. A rater has to be able to answer the question "why is this number this number" months or years after the fact, to a carrier, an auditor, a reinsurer or a regulator.
Where ISO content stops and the insurer starts
This is the distinction people find most useful, and it is the one that explains why two carriers using the same bureau content quote the same risk differently.
The bureau layer is licensed content: advisory loss costs, classifications, territory definitions, rules and the editions they belong to. It is maintained and reissued over time, and an insurer does not author it.
The company layer is the insurer's own position: loss cost multipliers that convert an advisory loss cost into that company's rate, deviations that differ by state, and the company's own approach to limits, minimums and eligibility.
The program layer, where one exists, is narrower still: the pricing and rules specific to a particular program or delegated authority, sitting on top of the company position.
A rater's real job is to keep those layers separate. When the bureau reissues content, the insurer's multipliers and rules should carry forward rather than be re-entered. When a company changes a multiplier, the bureau content underneath should not need to be touched. Systems that collapse the layers into one set of hard-coded numbers work exactly until the first circular arrives.
Who uses one
Not everyone who sells commercial insurance needs a rater of their own. The organizations that do are the ones that carry rating responsibility rather than just receive a rate.
Carriers
Own a filed position and need it applied consistently by everyone writing on their paper.
MGAs
Hold delegated authority, often from several carriers at once, and have to rate the same submission against each carrier's own position.
Wholesalers and program administrators
Distribute to retail agents who are not their staff and need the rate those agents quote to be the current one.
Retail agencies writing small commercial through carrier portals typically rate through someone else's system rather than licensing and running an ISO rater themselves. They consume rates produced by somebody else's.
Effective dates, and why they dominate the design
A commercial policy is rated on the content and the pricing that were in force on its effective date, not on what is in force today. Bureau editions change. Company multipliers change. Program rules change. A policy written in March has to keep re-rating in March's terms for as long as anyone asks about it.
That single requirement shapes most of what a serious rater does. Configuration has to be versioned by effective date rather than overwritten. Prior versions have to stay callable rather than archived. A quote requested for a date where a configuration is not yet effective should return no premium and say why, rather than quietly using whatever version happens to be current.
It is also the reason "we exported the rates to a spreadsheet" is a state most organizations pass through once and then leave. A spreadsheet has one version, and it is whatever was last saved.
What a good one produces besides a number
A worksheet. The rating as the engine executed it: each step, the values that governed it, and the rule that fired. Not a summary written afterwards, but the record captured at the time.
Provenance. For values the system derived rather than received, a statement of what it selected and why. A territory assignment that records which filed mapping governed, and where it came from, can be checked. One that just names a territory and stops there cannot.
Reproducibility. The same quote, re-requested years later, returning what it returned the first time, because it resolves the version and edition it was originally written under.
A clear failure. When rating cannot produce a defensible number, the answer should be no number and a reason. Every premium a system produces is one somebody may have to stand behind.
How rating fits with everything else
A rater is not a policy system, and the good ones are explicit about that boundary.
Rating prices a risk and records how. A policy administration system owns the customer record, billing, documents and accounting. Where the line sits varies by product: some raters stop at the quote, and some, including SelectRate, also rate and record the transactions on the policy term (endorsements, cancellations, reinstatements, renewals) while leaving administration to the system that owns it. A rater that is honest about where it stops is easier to integrate with whatever you already run.
In practice that means a rater is reached two ways: by underwriters working directly in it, and by other systems calling it over an API so that a portal, an agency system or a policy system can get a rate without a human retyping the risk.
Common questions
Is an ISO rater made by ISO?
No. ISO, part of Verisk, publishes the advisory content. Rating software that uses that content is built by other companies who license it. SelectRate is built and operated by Selectsys Tech LLC and is a licensed user of ISO content.
Is an ISO rater the same as a comparative rater?
No, though the names are close. Comparative raters are mostly a personal lines tool for retrieving quotes from several carriers for the same consumer risk. An ISO rater calculates commercial premiums from bureau content and company pricing, and is typically operated by the party that owns the rating position.
Does using ISO content mean every carrier charges the same?
No. Bureau loss costs are advisory. Each company applies its own multipliers, deviations, minimums and rules, which is exactly why the same commercial risk produces different premiums from different carriers.
Can an ISO rater be used through an API?
Yes, and for most organizations that is the point. The rater becomes something you can hand to a portal or a partner rather than a screen only your own staff can reach.
Which commercial lines does an ISO rater cover?
It depends on the system. SelectRate rates the four primary ISO commercial lines, general liability, commercial property, businessowners and commercial auto, and ten AAIS products.
Disclaimer: ISO is a registered trademark of Insurance Services Office, Inc. SelectRate is a licensed user of ISO content and is not affiliated with, endorsed by or sponsored by Insurance Services Office, Inc. or Verisk Analytics, Inc.
See one working.
The quickest way to understand a rater is to watch a real program get configured and rated. Bring one you know well.