Use case · Commercial and speciality

From broker submission to a price you can talk through

Fleet, property, liability and speciality risks don't arrive as tidy forms. Claude or ChatGPT reads the submission, Swallow prices it with your rating model, and underwriters work through the options in the conversation

Talk to our team
Claude and ChatGPT over MCP

Documents in, a price to work with out

The broker's pack goes into the chat. Claude fills in the model's inputs and Swallow prices and saves the quote. Ask for a what-if and it's priced on the same model. Ask why it was referred and you get the rule and every factor behind the price.

The prices here are real output from a commercial fleet model on the Swallow engine.

What it takes

Underwriting judgement, without the rekeying

Commercial pricing depends on people. Swallow takes the slow parts off them and leaves the judgement with them

Unstructured
Submissions aren't forms
Risks turn up as broker emails, vehicle schedules, statements of values, loss runs and proposal forms. Someone has to turn them into rating inputs.
Judgement
The underwriter makes the call
The model gives a consistent technical price. The underwriter or broker decides what to offer, within their authority.
Explainable
Every price has to hold up
Brokers ask why a price moved. Managers, auditors and capacity providers want to see exactly how it was built.
Time
Fewer risks, more work each
Each risk is worth the effort, but no one wants to spend a morning rekeying a 40-vehicle schedule before they can quote.
How it works

Read, price, try, explain

Build your rating model in Swallow from your rating guide, rater spreadsheets or filings, or with our team. Underwriters then use it from the AI tools they already work in

01 · AI client
Read
Drop the submission into Claude or ChatGPT. It fetches your model's quote form from Swallow, fills the inputs from the documents and tells you what's missing, so you check inputs rather than type them.
02 · Your rating model
Price
Swallow runs your published model and returns the technical price with any refers and declines, and saves the quote under your reference. It's the same engine that prices high-volume products.
03 · In conversation
Try what-ifs
Ask in plain English: a higher excess, telematics fitted, young drivers removed, a different limit. Each one is priced by the model and saved, so you can work through options with the broker.
04 · Full trace
Explain
Any quote can be opened step by step: every factor, every formula and every rule that fired. The underwriter takes the decision with the whole calculation in front of them.
Every quote saved

A full record for every risk

Every quote and what-if is saved with its inputs, result, model version and your reference. Open any of them for the full trace, from the dashboard, the API or an AI client, and see exactly what the model did
Two ways to price

Commercial or high volume?

One engine prices both. What changes is where the inputs come from and who makes the final call

Commoditised, high volume

Personal lines and embedded cover
See how →
Inputs
Structured, from forms and APIs
Who decides
The model's rules, with no one in the loop
The price
Final, and the same every time
What's kept
Every quote, with its inputs and model version

Commercial and speciality

Fleet, property, liability and speciality
You're here
Inputs
Broker emails, schedules and loss runs
Who decides
An underwriter or broker, working in an AI client
The price
A technical price for the underwriter to work from
What's kept
Every quote and what-if, each with its full trace