Intellect Design Arena Ltd · NSE:INTELLECT
Intellect Design Arena — Investor Presentation (July 2019)
35 slides from Intellect Design Arena's investor presentation, Jul 30, 2019
Slide 1: intellect SEEC
1 intellect SEEC www.intellectseec.com
Slide 2: Focus: Data
2 FOCUS : DATA intellect SEEC www.intellectseec.com
Slide 3: Why Data?
3 Why Data ? USD Millions US Commercial Insurance Market $314,925 US Commercial IT Spend (@4%) $12,597 Estimated US Commercial Packaged Software spend $1,575 vs. Verisk insurance revenue $2,440 Transunion $2,317 intellect SEEC www.intellectseec.com
Slide 4: Why Data?
4 Why Data ? intellect SEEC Source : IDC www.intellectseec.com
Slide 5: The Intellect Journey
5 The Intellect Journey ML Analytics ML Automation Cloud Native SW, Platform As a Industry Impact Service, Fabric Data Service Big Data Platform Augmented UW Cloud (non multi tenant) Enterprise SW Multi Channel, Java Mobile Software 2014 2015 2016 2017 2018 2019 Demerger Time intellect SEEC www.inte
Slide 6: Intellect is making investment in Tech an appreciating asset for Insurers
6 Fundamental Promise Intellect is making investment in Tech an appreciating asset for Insurers New Age ML Software High Dimensional That Automatically Streaming& Big Data Algorithms for core functions Evolves and Improves intellect SEEC www.intellectseec.com
Slide 7: What do we do?
7 WHAT DO WE DO ? Create Magic With Data, Analytics and Software Data Analytics Cloud Software Platform designed to be default Partner to build, test, deploy Cloud Native SAAS for Intake source of all 3rd party data : and continuously refine of commercial Assimilate new data sources predictive model
Slide 8: Business Model
8 Business Model H i g h l y S c a l a b l e , A d d i c t i v e a n d Va l u e G r o w i n g Data Analytics Cloud Software • API Based • Per use license model to • SAAS Subscription • Per Click (w minimum grow with scale • Attractive entry level annual subscription) • Creating our own pricing • Dif
Slide 9: Client Journey
9 Client Journey Analytics Loss Control Predictive UW j Prospecting Data Client GL Auto Prop Target Value & j TAM WC Addictiveness US$5Bn. Software US+UK+CA Xponent Intake – New LoBs P&C Xponent Intake Xponent Work Station Time How IT & Process What decisions How decisions Is Managed are made on are
Slide 10: Data world is exploding and enabling game changing use cases
10 Data world is exploding and enabling game changing use cases Forklifts/Lathes IoT/IoE (GPS, Giro, Fork Risky/Alexa position, Video, Collision Voice/ChatLSTM/CRF) data, Faults, Mechanics) Push Notifications Employees Commercial Driving License, Accident History, Health Old School Data (Smoking, Ar
Slide 11: The Industry has a big Data Problem
11 The Industry has a big Data Problem B i g D a t a & A I S h o u l d b e t h e a n s w e r, b u t t h e r e a r e c o m p l i c a t i o n s 1. User submitted data, particularly from brokers is messy, inaccurate and biased against the carrier 2. Data quality from traditional sources is inaccurate,
Slide 12: Enterprise Big Data Platform
12 I n t e l l e c t FA B R I C Enterprise Big Data Platform Chat bots No Fill UI Distribution Data Enrichment Risk Analytics Intellect Fabric Data Service Underwriting Search > Match > Aggregate > Contextualize > Analyze Predictive Rating Fraud Investigation Operations Customer Feedback Intelligent
Slide 13: Using Data to make magic: Insight + Foresight
13 Using Data to make magic : Insight + Foresight • The trick is not that the app was able to trigger a some rule: if get_from_api(destination_close) < 1hr alert(“Blah Blah”) • The trick was to infer you will arrive at 20:12 based on • Weather conditions • Road conditions • Historical traffic patter
Slide 14: Use Data to make magic: Insight + Foresight
14 Use Data to make magic : Insight + Foresight intellect SEEC www.intellectseec.com
Slide 15: A picture is worth a thousand words…But…
15 Computer vision A picture is worth a thousand words…..But… …does your system know how to read? intellect SEEC www.intellectseec.com
Slide 16: …it can with Deep Learning!
16 Computer Vision …it can with Deep Learning! roof (0.97624326) people (0.9719751) family (0.9543848) man (0.9538991) house (0.9445791) construction worker (0.930442) roofer (0.9325291) adult (0.9107766) action (0.9062755) business (0.9005916) safety risk(0.97167285) rooftop (0.92714784) Image cred
Slide 17: The Intellect SEEC Difference?
17 The Intellect SEEC Difference? Best and Most Insurance Data in the World 1. Velocity Search based - Most recent and accurate results 2. Variety Private data, open data, surface web, dark web, social and streaming 3. Veracity Match the right entity + Triangulate the element of truth from multiple
Slide 18: 8000+ Sources from Structured to Unstructured
18 Va r i e t y 8000+ SOURCES FROM STRUCTURED TO UNSTRUCTURED Company Location Person Unstructured 120 22 13 8,155 Company Location Person Violations Hazards Vehicles Owned Ratings Toxic Sites Social Info Industries Geo Info Employment History Financial Images Videos Driver Information Recognitions
Slide 19: Vast and Rapidly Growing Library of Raw & Triangulated Data
19 Vo l u m e Vast and Rapidly Growing Library of Raw & Triangulated Data and Predictive Insights Aggregation (Information) Data Value Chain Insights (Knowledge) Standard Triangulated Triangulated Rule Based Big Data Roadmap Grand Total Data Data Score Insights Insights A - Company 324 818 77 3 23 1
Slide 20: API Based Data Service Model
20 API Based Data Service Model Company Location Data Element ID Data Element Data Element ID Data Element C100011 Years in Business (Start Date) L100003 Location Address C100309 Legal Names L100004 City C100167 Legal Status L100005 Zip C100002 Annual Revenue L100007 State C100006 # Employees L10003
Slide 21: Intellect Data: August Release
21 Disruption of Traditional Data Monopolies Intellect Data : August Release Intellect Delinquency Score Underwriters can currently get information on past delinquent payments. But there has always (C101329): been a concern regarding predicting the future occurrence of missed or late payments. Intel
Slide 22: AI Opportunity - Automation Spectrum
AI Opportunity - Automation Spectrum 22 intellect SEEC www.intellectseec.com
Slide 23: "Data First" strategy in automation
23 “Data First” strategy in automation 800+ >90% 40-50% >20% new risk and exposure data elements more data Improved data automation available for coverage of forms accuracy in key potential better risk than brokers risk elements selection, underwriting & rating Reduce Cost Improve Data Fill Increase
Slide 24: IDX Video
24 IDX Video intellect SEEC www.intellectseec.com
Slide 25: IDX Video
25 IDX Video intellect SEEC www.intellectseec.com
Slide 26: GTM Evolution
26 Opportunity Space GTM Evolution Verticals Use Cases Channels 1. US Commercial Insurance 1. Data Validation 1. Complex Enterprise Sales Pilot 2. Simple Enterprise Sale 2. Secondary markets - 2. Data Enrichment 3. Market Place Pilot Pilot Q3/ Q4 commercial Insurance • Junto Phase • AWS Marketplace
Slide 27: Recent US Insurer Case Studies
27 Recent US Insurer Case Studies intellect SEEC www.intellectseec.com
Slide 28: Finding right prospects and prioritizing to top of queue
28 Example 1 : Sharpened prospecting and Prioritization Finding right prospects and prioritizing to top of que Assumed Prospects Rating Real view of Portfolio with Intellect Data Before Intellect Data ~75% Best Risk ~15% Investing resources in the wrong opportunities due to ~40% 25% classified incor
Slide 29: Scan book in minutes for key risk elements
29 E xa m p l e 2 : D a t a f o r I n s i g h t s a t S c a l e a n d Ve l o c i t y Scan book in minutes for key risk elements • Extract data deep within public sources (OSHA/SAFER) • > 1,300 elements on every account • Apply NAICS/SIC rules for eligibility with upto 20+ codes including reverse eng
Slide 30: Correcting erroneous input data for key Rating and Underwriting fields
30 E x a m p l e 3 : D a t a Va l i d a t i o n Correcting erroneous input data for key Rating and Underwriting fields Client Example : Classified as food services Vs. actual business is butchery Standard Data : SIC 2011 - Meat Packing Plant (general) Big Data Insights : Slaughtering/butchering (ide
Slide 31: Text Mining to Accurately Answer key UW Questions: Liquor Example
31 Example 4 Text Mining to Accurately Answer key UW Questions : Liquor Example Show relevant & meaningful data to users Y/N text alert (simple UW alert) Unstructured Data collected Expert standardized context (coaching and insights with recommended interventions) intellect SEEC www.intellectseec.co
Slide 32: Cut Manual Underwriting Effort from hours to minutes
32 Example 5 Cut Manual Underwriting Effort Cut from hours to minutes Quantitative Benefit : Automation of Manual UW Effort Qualitative Benefits Underwriter searches of 2 hours per account on source research Consistency among UW staff Finding nothing in 30% of the cases Expert level guidance
Slide 33: Industry Accolades
Industry Accolades 33 Major analyst Recognitions Major Industry Awards & Recognition Intellect SEEC's among Early Adopters of AI in Insurance Underwriting features. Risk Analyst is one of the leading Predictive Analytics Solutions for US Insurers. Risk Analyst rated the strongest in its ability to s
Slide 34: intellect SEEC www.intellectseec.com
34 intellect SEEC www.intellectseec.com
Slide 35: intellect SEEC www.intellectseec.com
35 intellect SEEC www.intellectseec.com
Intellect Design Arena — Investor Presentation (July 2019) on 6ix