TDI Startup Insurtech Award - Riskine presentation
1. riskine
I n t e l l i g e n t C u s t o m e r A d v i s o r y &
I n d i v i d u a l C u s t o m e r J o u r n e y s
0 7 / 2 0 1 9
A . C O M PA N Y O V E R V I E W
B . I N T E L L I G E N T C U S T O M E R A D V I S O R Y
C . I N D I V I D U A L C U S T O M E R J O U R N E Y S
2. 2
• DeepTech made in Vienna/Austria
• B2B Advisory AI (White Label)
• 10 leading banks & insurers (implemented: Generali, VIG, Erste Bank, Swiss-
Life, Merkur, ÖBV, Wüstenrot, TIS/CCA, wefox, Allianz, TIS, CCA..)
• Current markets Austria, Germany, Switzerland, Spain and Japan in Rollout
(SBI)
CREATING INTELLIGENT SOLUTIONS
FOR DIGITAL ADVICE
We are proud to be one of Gartner’s "Cool Vendor 2018". The prestigious
"Cool Vendors in Insurance 2018" report analyses upcoming technologies which provide innovative
products and services within the insurance industry. riskine was selected in this year's report as one out
of four businesses worldwide.
A. COMPANY OVERVIEW
3. 3
riskine FINANCIAL ADVISORY AICUSTOMER
• Which insurance do I need ?
• Which investments should I conduct?
• Can I afford a house?
• Should I buy or rent?
• Should I buy or lease a car?
• How much should I save per month?
• Which credit card?
• Continuous, real time, event driven
• Personalized – knows and remembers
me and my preferences
• Wholistic – life time cash flow prognosis
per person/family
• Understands me - Asks back & conducts
grounding
• Multichannel: text, voice
• Smart – Monitors the web for me
?? ?
?
VISION: EVERY INDIVIDUAL TAKES
SOUND FINANCIAL DECISIONS
B A SED ON OU R A D VISORY A I
A. COMPANY OVERVIEW
4. 4
ADVANTAGES RISKINE SOLUTION
BETTER CUSTOMER EXPERIENCE
• Better Advisory
• Fast and Interesting Customer Journey
MORE TURNOVER VIA HOLISTIC ADVISORY
• Holistic advisory algorithm leads to better advice and higher Cross Selling
(+20%)
MORE INTELLIGENCE AND DATA – AI BASED
• Every step in the customer journey is used to generate data
• Data is processed in real time: adjust path, product recommendations
EFFICIENCY/COST SAVINGS
• Solution minimizes number of questions to customer (time saving)
• Savings on advisor trainings via built in background information/knowledge
graph
• Possibility to reduce headcount in sales
MULTICHANNEL & API BASED
• Solution is built for omnichannel usage- Cross channel usage of data
A. COMPANY OVERVIEW
5. 5
INTELLIGENCE BASED
ALGORITHMS & DATA
• Algorithms & Machine
Learning modules for
Demands and Needs
analysis and product
recommendations
• Ontology Graph & Ontology
holding the domain specific
knowledge
• Databases to reduce
required number of
questions and improve
advice quality
USER DATA
ML BASED PRODUCT
RECOMMENDATION
B. INTELLIGENT CUSTOMER ADVISORY
DATA BASES CRAWLED DATA
KNOWLEDGE GRAPH &
ONTOLOGY
ADVSIORY
ALGORITHM
6. 6
ADVISORY ALGORITHM OVERVIEW
Calculation Logic
1. Probability of 8 life risks based on statistical
data and individual answers
2. Need of protection across 8 life risks
a. Calculate all potential maximum
damages/losses over lifetime
b. Deduct all eligible state benefits
c. Weight above losses with risk probability
and
aggregate towards 8 life risks
3. Insurance policies are prioritized according
to the maximum claims which could arise
and the ability to cover these claims
B. INTELLIGENT CUSTOMER ADVISORY
Risks probability & Protection Relevance
Product Prioritization
7. 7
MODELLING DIGITAL DIALOGUES
WITH GRAPH DATABASES
Complex question &
answer flows
Question
Question
Example: Advisory processes for bancassurance
• Complex dialogues with hundreds of questions & thousands of connections
• “Bank advisors are guided through advisory and receive supporting background
information for customer and x-selling leads”
• Each customer gets customized questions and question-paths (=dialogues)
C. INDIVIDUAL CUSTOMER JOURNEYS
8. 8
• Graph steers the overall conversation flow
• Questions combined to a dialogue flow via
rules
• Natural language understanding is driven by
a neuronal network that allows open
questions
• NLU integration allows deep (e.g. product
advice) & broad conversations (FAQs)
2. NLU1. DIALOGUE FLOW
COMBINING COMPLEX
CONVERSATION FLOWS WITH NLU
C. INDIVIDUAL CUSTOMER JOURNEYS
9. 9
ALLOWING BROAD &
DEEP DIALOGUES
I N T E G R AT E D N L U T E C H N O L O G Y
• Usage 1 – Question type
Example: Which pet do you have? Answer: Labrador, System: recognises dog
• Usage 2 - Dialogue management: Jumps to the relevant conversation part
Example: What can I do for you? Product advisory, general demands and needs analysis, claims, services, contact,..
FAQ
PRODUCT ADVISORY
DEMANDS AND NEEDS
ANALYSIS
Deep Dialogues
BroadDialogues
C. INDIVIDUAL CUSTOMER JOURNEYS
10. 10
TECHNOLOGICAL ADVANTAGES
Optimal questions for maximal customer
satisfaction
reduce total number of questions and adapt
questions to the customer
(shortest path query through the graph)
C. INDIVIDUAL CUSTOMER JOURNEYS
Individual text labels
various languages, long text vs short text
labels (e.g. for chatbot), added info-texts
11. 11
ANALYTICS
Customer journey/data related
analytics:
• Aborts
• Time per question
• Customer profiles
• Clusterings
Ontology related analytics: for
example semantic structure of
question and paths
MICRO MEASURING THE CUSTOMER
JOURNEY
C. INDIVIDUAL CUSTOMER JOURNEYS
Customer answers are linked to relevant question –
automated clustering
What is your
gender?
10 s 2.5 m 4 m
1.5 m
15 s
2 m
15ss
0% 0% 1%
3%
1%
1%
0,5%
Time per question
Churn per question
12. 12
12
Dr. Ralf Widtmann
CEO riskine
+43 664 840 7618
Ralf.Widtmann@riskine.com
Frederik Schorr
COO riskine
+43 676 9054664
Frederik.Schorr@riskine.com
riskine GmbH
Waaggasse 15/1
1040 Vienna
Austria