This document discusses how insurers can modernize and innovate using data and machine learning. It suggests that insurers combine traditional and non-traditional data sources, like telematics and social media data, and use machine learning to better understand customers, detect fraud, predict claims, and optimize pricing. The document also recommends that insurers partner with other organizations and develop insurance apps to take advantage of new data sources and analytics tools to improve customer experience, compliance, pricing, and reduce costs.
3. What keeps insurers awake at night?
Client Understanding
• Can Insurance companies do a better job of identifying and valuing
their better and worst customers?
• How can Insurance companies innovate using media and other
communication channels to acquire new customers or deepen their
relationship with the existing ones?
Strategy & Growth
• How will changing consumer socio-economic and
demographic forces impact for Insurers products?
• How will key macro economic and regulatory changes
impact growth and opportunities ?
Customers
Regulation
Economy
4. What keeps insurers awake at night?
Fraud mitigation
• One of the biggest areas where insurers suffer of
enormous expense line.
Sales & Distribution
• How should Insurers companies improve the customer
experience through each distribution channel to maximise
sales and profit?
• Can be the pricing model optimised by capturing new data
to apply to underwriting process?
Crime
5. Modernisation Process
Change of mind set is required
• From the legacy technologies used towards new emerging
technologies
• From a claims leakage process that is reactive to one that
is proactive—potentially leading to enormous potential
savings.
6. 6
Combining the data sources
Traditional data Non-traditional
Unstructured
Web Sentiment
POQ - POS - MTA - FNOL
Customer declared data
Emotional
context
Linked
Addresses
Intre
Integrate
Analyse
Visualise
Discover
Credit
Bureau
External data
IDV Claims
History
Vehicle Fraud
Disparate data
Legacy systems
Telematics
7. Changing the paradigm
The Unknown
Previously unknown
metrics revealing underlying trends
and patterns driving new questions.
The Known
Rapid multilayer analysis utilizing
big data analytics techniques.
9. Machine Learning - Predict!
Questions for the ML
• Will this policy holder have an accident?
• How much will be the refund of a given claim?
More complex
• Given telematics data about two trips in different cars, can you
say that the driver is the same?
Automatic design
models from data
If you can automate the
reasoning behind a model built
by a human you can replicate
his effort as many times as you
want and with a much smaller
amount of time.
.
10. Innovating with our partners..
In association with…….
“ecosystem”
Insurance apps
Unstructured
Images
Telematics
Traditional dataClaims
Vehicle
Credit
Fraud
Locational
Revolutionary Analytics tools
Device led data
Psychometric
Profiles
Telematics
+
• Internet identity
• Pre-validated profile
Rich Data Sources
• Intentions
• Hopes
• Fears
• Feelings
• Telematics
11. We shall deliver….
Many more hidden insights from existing data! to drive previously unasked
questions…………...
Legal
entitlement
DPA section 7
Improved customer on-boarding journeys.Commercial
Digital passportTelematics
Fully Compliance Services
More accurate dynamic pricing
Real time scoring techniques
Data visualisation
Operational Improvement to reduce human failure
Pattern Detection on Customer Behaviour
Improved Customer service by KYC analytics
Emotional ContextCrif Footprint
More and ore Insurance companies are realising the benefits of using advanced analytics for desiging products , developing strategies and metrics for risk management
Now The question is:
AS CRIF Insurance Competence Center What do we have in mind for the next years?
Innovate using a broad range of traditional and advanced techniques to generate insights.