Customer intelligence for insurers, beyond CRM
The software that tells you which policyholders are profitable, which are likely to leave and what to offer them, installed at the insurer.

Application screenshot, fictitious data
What is Moost Assurance?
Customer intelligence software, installed at the insurer. Connected to your existing data, it tells you which policyholders are profitable, which are likely to leave, what to offer them, and what each action brings in.
What a policyholder costs in claims, compared with what they pay in premiums. The lower the loss ratio, the more profitable the policyholder. Moost calculates it for every policyholder, over 3 years.
Each point of profitability is worth 1% of your premiums
One point less of loss ratio, or one point more of retention. Moost shows where to gain these points, policyholder by policyholder, and measures the gain against a control group.
Spot profitable policyholders about to leave
Every policyholder is placed by profitability and departure risk. Each cell has its recommended action, editable by your teams.
Example: 13,800 highly profitable policies at high risk, the agents' priority.
Act before renewal, not after
The 10 playbooks follow the life of the policy. Renewal is at their heart.
Departure score
The non-renewal risk is calculated for every policy.
Agent call
Profitable policyholders at risk enter the call queue.
Final reminder
Reminder under a contact cap, through the preferred channel.
Measurement
Renewal rate compared with the control group.
Every score is explained, policyholder by policyholder
In red, what raises the risk; in green, what lowers it. The agent knows what to say before calling.
Moost Assurance, screen by screen
Click a screenshot to enlarge it.
The portfolio at a glance
Active policies, 3-year loss ratio, retention and multi-policy, month on month.
- Policyholders by profitability segment
- Lifecycle stage
- Actions awaiting approval
Reliable scores, under control
Non-renewal at D-90 and multi-policy propensity at 180 days.
- Minimum quality of AUC 0.65 before going live
- Retrained every month, approved by a person
Describe your audience in plain language
"Policyholders in the South zone driving an SUV, renewal in 30 days or less": 1,418 policyholders found, ready for an action.
- Days to renewal, vehicle type, risk zone
- Editable filters, export or action in one click
Safe actions, approved by your teams
Before anything is sent: exclusions (including policies cancelled for non-payment), consent, contact cap, control group and human approval.
- The number of policyholders removed at each step is displayed
- Sent by file, signed webhook or API
The gain is proven, in premiums saved
Renewal at D-60: 81.2% among treated policyholders versus 72.4% in the control group, i.e. +8.8 points and TND 412,000 of additional revenue.
- Return on investment shown as soon as the cost is entered
Connected to your data, read-only
Six sources: policyholders, policies, premium notices, claims, renewals, agents. Through files, Oracle, SQL Server, PostgreSQL, MySQL, MariaDB databases, SFTP or API.
- One policyholder, one profile: matching by national ID, customer ID or name, city and date of birth
CRM manages the relationship, Moost drives profitability
Moost does not replace the CRM: it feeds it the right policyholders, at the right time, with the reason for the action.
"Who is this customer, and what happened with them?"
"Which policyholders to act on, why, and what does it bring in?"
| CRMDynamics 365, Salesforce, HubSpot | Microsoft Customer InsightsCustomer data and journeys | Moost AssuranceInsurance customer intelligence | |
|---|---|---|---|
| Segmentation | PARTIAL Lists and filtered views | YES Rule-based segments, with Copilot | YES Insurance segments, plain-language search |
| Loss ratio per policyholder | NO Premiums and claims outside the CRM | NO Indicator to model | YES Over 3 years, 4 segments |
| Cancellation risk | NO Lead and opportunity scores | PARTIAL Generic churn | YES At D-90, explained per policyholder |
| Multi-policy | NO No model | PARTIAL Recommendation from purchases | YES Propensity at 180 days |
| Actions and campaigns | YES Tasks and campaigns | YES Multichannel journeys | YES 10 insurance playbooks |
| Proof of gain | PARTIAL Clicks and conversions | PARTIAL Journey engagement | YES Control group, premiums saved |
| Data and AI at the insurer | PARTIAL Cloud; Dynamics 365 on-premises until 2029 | NO Azure cloud, Copilot in the cloud | YES Your servers, local AI |
Installed on your premises, built for sensitive data
On your servers
No data leaves your network. Installation possible without internet access.
Protected personal data
Identifiers replaced by a hash, personal data encrypted, export and erasure on request.
Controlled access
One-time code or company directory, four roles.
Local AI
Models and generative AI on your servers, with no personal data transmitted.
Traceability
Chained, verifiable audit log of every action.
Operations
Tested backups, updates with automatic rollback.
Start with a pilot line of business
Data workshop
Inventory of sources and choice of the pilot line, for example motor.
Installation
On your servers, with 3 years of history and the loss ratio of every policyholder.
Scores and playbooks
Models trained on your data, two or three playbooks launched.
Measurement and decision
Premiums saved against the control group, then the decision to scale.
Perpetual licence for the Assurance edition and an annual maintenance contract. Proposal based on your volumes and scope.
Born from a project with BH Assurance
Moost Assurance comes from a proof of concept run with BH Assurance, in motor insurance.
Frequently asked questions
1. Does Moost Assurance replace our CRM or core insurance system?
No. Moost reads your policies, premium notices and claims read-only and sends agents and the CRM the right policyholders, with the reason for the action.
2. What data do you need?
Ideally 3 years of history: policyholders, policies, premium notices, claims, renewals and agents, through files, database or API.
3. Does our data leave the company?
No. Moost Assurance is installed on your servers, the AI runs locally and installation is possible without internet access.
4. Can agents understand the scores?
Yes. Every score is explained by its main factors, in plain words, on the policyholder profile. The agent knows what to say before calling.
5. How do we get started?
With a pilot on one line of business, for example motor: data workshop, installation, two or three playbooks, then a decision on premiums saved.