Draft comparison explanations and records
When a multi-insurer agency recommends a product, AI drafts the quote comparison and the reason. Staff edit it before explaining it and keeping the record.
- Runs on your own PCs and servers
- Use Azure OpenAI, Amazon Bedrock, Google Vertex AI, or an in-house model
- Its engine is Descarty's open-source Dagu (GPL-3.0)
24 quotes from 3 insurers for 8 clients read and compared in 80 seconds, with a comparison table and a reason drafted for each; for 6 of the 8, the lowest-priced quote missed the client's wishes, so one that met them was recommended. A person approved and the records were saved. Kitewell built the workflow 10 minutes after the request.
A real run on October 28, 2026, sped up. The system it works with was built for the demo.
The problem today
When a multi-insurer agency compares and recommends products, Japanese rules require explaining the comparable products and the reason for the proposal. Quotes are made on each insurer's own screens, and staff compile comparison tables and records by hand. Records vary by who wrote them and take time to review.
The workflow in Kitewell
- Add each insurer's quote to a batch sheetAutomatic
- AI reads coverage and premiums from the quotes into a comparison tableAI judgment
- AI drafts the reason for the proposal from the customer's needs and the comparisonAI judgment
- Staff check and edit the table and reasoning, then approveApproval
- Save the explanation and record in the CRMAutomatic
What AI does
Reads coverage and premiums from quotes, quoting the part of the quote each value came from, and drafts the reason for the proposal in light of the customer's needs.
Where people decide
Staff decide which product to recommend. They check and edit AI's draft before using it to explain and record.
Prove Kitewell on your own systems
Try it free for 14 days on your own PC. To see it working first, ask for a 30-minute demo. There is also a one-month onboarding (¥500,000 before tax), where we build your first job together.
What to check when you try this work
- Whether coverage and premiums are read correctly from each insurer's quotes
- Whether the drafted reasoning follows your rules for comparison-based recommendations
- Whether the record's location and format fit how you manage records
How the onboarding runs
- Week 1Requirements and success criteria
- Weeks 2 to 3Build
- Week 4Verify on production screens
- Final dayReport
One-month onboarding
- Length
- One month
- Cost
- ¥500,000 (tax excluded). Your staff build, with the developer beside them
- If you go on to a license
- The onboarding fee is credited to the first year's contract
- Without the onboarding
- At Team scale, start on Team monthly (¥14,800 per computer a month, tax excluded), and cancel in any month
You can try it first on your own PC, in the free 14-day trial.
Frequently asked questions
- Do our documents and screens leave the company?
- Kitewell runs on your own PCs and servers, and Descarty never receives your business data. You choose the AI model: a cloud API, or an in-house or local model with an OpenAI-compatible API. With an in-house model, no data leaves the company. With a cloud model, what the AI reads is sent to the model's provider.
- What if the AI misreads something?
- A person checks what the AI produced before the work goes on. In a sheet, every value the AI reads shows the quote it came from, and rows below your confidence threshold go to a person. Reviewers can send work back with notes on what to fix, and who approved what is kept in the run history.
- What happens when a screen changes or a run stops?
- When a screen changes, the AI reads it again, relearns the step, and carries on. The relearned step is saved, and later runs repeat it. If a run still stops, ask the assistant right there, and it finds the cause and proposes a fix. A proposal is not saved until a person applies it.
- What happens to the RPA and systems we use today?
- Keep them. Kitewell operates screens the way a person does, so it works with systems that have no API. Move the robots that keep stopping and the work you gave up automating because it needs judgment, and leave fast, fixed processing in your current RPA.
- What does it cost?
- ¥14,800 per computer per month, or ¥148,000 per computer when paid yearly, before tax. A computer costs the same whether it builds or runs. You can try it free for 14 days. The one-month onboarding, where we build your first job together, is ¥500,000 before tax, credited to the first year if you go on.
Related use cases
Find expiring policies and send renewal notices
Collect expiring policies through the insurers' shared gateway or agency systems, and draft renewal notices showing what changed. Coverage changes are flagged.
Get quotes from several insurers
Enter one set of conditions in each insurer's agency system to get quotes, and keep each premium with the conditions screen. Staff check items hard to map.
Track exclusive listing and REINS deadlines
Track REINS registration, certificate delivery, seller reports, and status-change deadlines for each exclusive listing. AI drafts reports; agents operate REINS.
See this work in a demo
A 30-minute demo shows this workflow running. The 14-day free trial lets you check it on your own systems.