Advisory Portal
For insurers, broker pools and sales teams
Needs analysis, plan comparison, health questions and application in one flow. The IDD advice record is produced along the way, not afterwards.
Needs analysis · Offer · Application · Contract
Brokers advise, customers sign, insurers receive complete applications. Every step is logged audit-proof and kept for ten years, in Germany and Switzerland alike.
Four building blocks for digital insurance distribution.
For insurers, broker pools and sales teams
Needs analysis, plan comparison, health questions and application in one flow. The IDD advice record is produced along the way, not afterwards.
For the back office
Reconciles the policies in your broker management system with the insurers’ data, enriches them and feeds financial analysis, form filling, customer journeys and mailing campaigns.
For end customers
After the consultation, customers take out the pension savings plan themselves: on their phone, with the details from the meeting, in their broker’s branding.
For broker pools and software vendors
Reads policies and endorsements from PDFs, validates them against business rules and delivers them as BiPRO 430.4.
AI consulting for insurers and financial services
An application with twelve attachments, a cancellation by email, a health questionnaire as a scan: cases like these cost insurers and sales teams hours every day. That is exactly where we put AI to work, in a way that data protection and the supervisor can stand behind.
From our own practice: Insurware itself is built with AI assistants, safeguarded by more than 100 automated quality gates.
The path of a decision
We look at the inbox, application review and policy administration and find the places where your team still types, sorts or chases details by hand. That is where we start, with a goal that can be measured.
AI proposes, a person approves. We define the confidence at which it may decide alone, when it escalates and how spot checks run.
GDPR and the Swiss FADP, DORA, the EU AI Act. We clarify which data a model may see, where it runs and how you show that it keeps to your rules.
A large cloud model, a small model in your own data centre, or both? We test with your real cases instead of benchmarks and work out the cost per case.
The model is rarely the problem; the integration is. We plan the interfaces to core systems, BiPRO and broker management software, plus a fallback for the day a model fails.
This is how we work ourselves: AI assistants write alongside us, automated tests hold the quality. We show your team how that works in a regulated environment.
Want to see a demo or discuss an AI project? Write to me or give me a call. I usually reply on the same working day.