Insurance agency automation: quotes, applications, renewals

Quotes, applications, renewals, and comparison records at multi-insurer agencies: re-entry into each insurer's system, automated with AI and human review.

  • 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)

Common problems

  1. Comparing insurers means entering the same conditions again in each insurer's agency system

  2. As insurers withdraw seconded staff and clerical help, applications and booking fall to the agency

  3. From March 2028, Japan's comparison-and-recommendation rules tighten, with more to explain and keep on record

A multi-insurer agency’s paperwork grows with every insurer it represents

Japan has 151,000 non-life insurance agencies, of which 36,000 represent more than one insurer (end of fiscal 2023). Agencies handle 89.7% of non-life premiums (March 2025).

A multi-insurer agency quotes, takes applications, books policies, processes changes, and sends renewal notices in each insurer’s own agency system. NTT DATA’s shared gateway for insurers lets agencies sign in to each insurer’s system with one ID and receive policy data. The screens for entering quotes and applications still differ from insurer to insurer.

Work Screens used What stays manual
Quotes Each insurer’s quote screens Entering the same conditions again for each insurer
Applications and booking Each insurer’s agency system Copying the application from the quote and booking the policy
Changes (address, vehicle replacement) Each insurer’s agency system Processing the change with each insurer a customer holds policies with
Expiry and renewal The shared gateway and each insurer’s system Collecting expiring policies and grouping them by customer
Comparison records Customer management system, Excel Writing up comparison tables and the reasons for a recommendation by hand

Quotes mean entering the same details again for each insurer

A fire insurance quote needs the building’s age, structure, floor area, and more, entered in each insurer’s own quoting system. IRRC Corporation, which sells a fire insurance comparison product, says this takes about 5 to 10 minutes per insurer, and comparing several insurers can take from tens of minutes to an hour.

In the public comment results the FSA published in August 2026, one commenter likewise said that because each insurer uses a different system to create quotes, getting and presenting quotes from several insurers takes considerable time and effort.

The FSA also replied that it does not uniformly require agencies to always create and present quotes from every insurer they represent. Staff decide which insurers to quote, following what the customer wants. What gets automated is the entry after that decision.

Insurers withdraw seconded staff, and the paperwork stays with agencies

As of March 2023, the four major non-life insurers had about 2,370 staff seconded to about 1,520 companies, 75% of them insurance agencies. Some agencies relied on those staff for insurance paperwork. The four insurers plan, as a rule, to end these secondments so as not to be suspected of improper favors to agencies.

The FSA’s supervisory guidelines for insurers, revised in August 2025, list an insurer providing services for work the agency should do on its own responsibility as an example of conduct that may be excessive favor. Yano Research Institute reports cases where agencies struggled with paperwork after insurers’ clerical support ended abruptly.

Quotes and applications that insurers used to help with are now the agency’s own work.

From March 2028, the comparison-and-recommendation rules change

On August 28, 2026, the FSA published revisions to the cabinet office ordinance and supervisory guidelines covering how multi-insurer agencies compare and recommend products. They take effect on March 1, 2028.

The revised guidelines ask whether, when recommending a particular policy, the agency explains criteria and reasons that are rational and reasonably specific, without regard to the convenience of the agent or the agency. Agencies must also set retention periods for records of comparison and recommendation in their internal rules, and check that the practice is appropriate.

The FSA states that merely presenting premium comparisons or a system’s output does not fulfil the duties to understand the customer’s intent and to compare and recommend. Automating quotes does not meet those duties. Working out the reasons for a recommendation and explaining them to the customer is the agent’s job.

The FSA’s 2026 insurance monitoring report also lists the burden of unifying comparison measures across insurers and adapting systems as an issue in comparison and recommendation. In addition, the revised Insurance Business Act, in force since June 1, 2026, requires large multi-insurer agencies to build internal structures for running their business.

What to check before automating

  • Whether automated screen operation is allowed: check your agency agreements and each insurer’s system rules. Nothing industry-wide settles it
  • Where customer information may be entered: in the FSA’s March 2025 action against Sompo Japan, multi-insurer agencies sending one insurer’s policyholder information to other insurers was treated as a data leak. Decide first where the information used for quotes comes from and how consent is obtained
  • How each system signs in: the shared gateway, client certificates, and one-time passwords differ by insurer
  • Who decides: staff decide which insurers to quote, which product to recommend, and what the application says. Items the customer must answer, such as declarations, are not automated
  • Whether a dedicated product is enough: dedicated quote-comparison products exist for the major non-life insurers’ auto and fire insurance. If they cover what you need, consider them first

What Kitewell does

Descarty’s Kitewell is integration and automation middleware that connects systems through APIs, files, databases, and screens.

  • Operates each insurer’s agency system: AI reads the screen and fills in quote and application screens. Operations that worked are remembered and repeated the same way on later runs
  • Keeps going when screens change: when an insurer’s system changes and a remembered operation no longer fits, AI reads the screen again, does the step, and remembers the new operation. Only if that fails does the run stop, and AI then looks into the cause and suggests the fix
  • Handles sign-in hurdles: for sites that ask for a client certificate, it presents the certificate registered on the computer for that site. When a one-time password is required, the run pauses until staff enter it
  • Passwords stay out of AI’s hands: agency system passwords are referenced by name and typed in without showing the value to the AI
  • People check where it matters: approval can be set on any step, such as confirming conditions or before submitting an application. Requests reach Slack, Microsoft Teams, and email
  • Every operation on record: each run keeps its logs, screenshots of the screens it filled in, and approval decisions, so you can later see what was quoted under which conditions
  • Runs from an Excel list: register a customer list as a batch sheet, and it runs row by row and writes each result back to its row

You can start from one piece of work, such as taking a quote request through getting quotes from each insurer, building the comparison, and sending the proposal. A demo shows how it runs, a free connection test confirms Kitewell reaches your systems, and a one-month PoC measures the result against success criteria agreed up front, before you license it for production. See From evaluation to production for details.

Sources

Frequently asked questions

May a tool operate insurers' agency systems?
It depends on your agency agreement and each system's rules. Nothing industry-wide says whether automated operation is allowed, so the PoC checks each insurer and automates only what its rules allow.
Does automating quotes meet the comparison-and-recommendation rules?
No. The FSA states that merely presenting premium comparisons or a system's output does not fulfil the duties to understand the customer's intent and to compare and recommend. The time automation frees up is meant for confirming what the customer wants and explaining the recommendation.
How does this differ from dedicated quote-comparison products?
Dedicated products already create quotes automatically for the major non-life insurers' auto and fire insurance. If they cover your insurers and lines, they are the quicker route. Kitewell suits quotes for insurers or lines those products don't cover, and joining quotes, applications, booking, and ledger updates into one flow that follows your agency's own steps.

Use cases for any industry

  • All industries
  • Accounting & finance

Read invoices into the accounting system

AI reads incoming invoices, checks them against purchase orders, and records them in the accounting system. Invoices whose amounts differ wait for approval.

  • All industries
  • Accounting & finance

Create invoices and send them by email

Invoices are filled into your template from an Excel billing list and emailed once a person approves. AI first points out amounts far from last month's.

  • All industries
  • Sales

Research a list of prospects

For an Excel list of hundreds of companies, each one's business, locations, and size are collected from its website into its row, with a priority for sales.

  • All industries
  • HR & payroll

Find missing clock-ins

Collect yesterday's clock-ins from the attendance system and, once HR approves, ask employees with missing entries to fix them in Teams.

  • All industries
  • Accounting & finance

Take in emailed invoices and draft the entries

Find unread email with an invoice attached, have AI read the amount, and record a draft ledger entry. Uncertain entries wait for accounting to confirm.

  • All industries
  • IT

Review SaaS accounts

A SaaS account review matches each service's users to current staff. Collect each console's users, match them, and disable leavers only after approval.

  • All industries
  • Accounting & finance

Partner registration that starts on approval

A kintone approval starts the workflow by webhook. It looks up the tax registration number, AI matches it to the request, and a match enters the sales system.

  • All industries
  • Planning & marketing

Collect KPIs from SaaS tools into a weekly report

Collect figures from advertising, sales, support, and other SaaS tools, and produce a weekly report with week-on-week comparisons and likely causes of change.

  • All industries
  • Accounting & finance

Check supplier invoice registration numbers

Check your supplier master's registration numbers monthly through the NTA's Web-API, and list suppliers that were revoked, lapsed, or differ in name.

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 the onboarding checks

  • Which method connects each target system: API, file, database, or screen
  • Whether Kitewell can work a system with no API through its screens, as a person would
  • Whether AI relearns a step and processing continues when a screen changes
  • Whether errors are handled as agreed: stop, notify, or retry
  • Whether it runs in your environment, with the AI model you choose
  • Whether processing time and AI usage cost fit daily operation

How the onboarding runs

  1. Week 1Requirements and success criteria
  2. Weeks 2 to 3Build
  3. Week 4Verify on production screens
  4. 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.

See how the onboarding runs and its terms