1 of 40

Smith Live Case BMGT495 | Presentation on 12/7

Group 4

1

2 of 40

Our Team

Kyle Covello

Junior | Finance & Info Systems

Julia Sepe

Junior | Finance & Info Systems

Caroline Fleetwood

Junior | Marketing & Supply Chain Management

Camille Peck

Junior | Marketing & Management

Kate Maescher

Junior | Marketing & Supply Chain Management

Mohamed Moneib

Junior | Finance & Info Systems

Amanda Schramm

Junior | Finance & Info Systems

2

3 of 40

Agenda

3

1. Our Purpose

3. Key Findings

4. Evaluation

2. Our Solutions

5. Next Steps

6. Conclusion

Our problem statement that we’re solving.

Our recommendations to address the problem.

Our methodology and findings that led us to our solutions.

Our pros and cons of each solution.

Our implementation plan of our final recommendation.

Overall review our problem and final solution.

4 of 40

4

Our Purpose

Our Solutions

Key Findings

Evaluation

Next Steps

Conclusion

5 of 40

Problem Statement

In today’s digital age, insurance customers want simplicity:

How can GEICO use technology or product innovation to gain a competitive advantage in the insurance industry. What are viable options for increased efficiency, or long-term sustainability?

5

6 of 40

6

Our Purpose

Our Solutions

Key Findings

Evaluation

Next Steps

Conclusion

7 of 40

Viable Solutions

7

1. Enhance Social

Media

Use AI technology to change Geico’s approaches to its social media platforms.

3. Increase Fraud

Prevention

2. Improve Claims

Management

Utilize AI to dramatically improve and enhance the process around Claims Management.

Implement AI technology to help further prevent insurance fraud.

8 of 40

8

Key Findings

Our Purpose

Our Solutions

Key Findings

Evaluation

Next Steps

Conclusion

9 of 40

Prevalence of Insurance Fraud

9

Digital Doughnut

10 of 40

Prevalence of Insurance Fraud

It is estimated that between $5.6 billion and $7.7 billion was fraudulently added to paid claims for auto insurance bodily injury payments in 2012, compared with a range of $4.3 billion to $5.8 billion in 2002. (IRC)
The total cost of insurance fraud (non-health insurance) is estimated to be more than $40 billion per year. That means Insurance Fraud costs the average U.S. family between $400 and $700 per year in the form of increased premiums. (FBI)
About 90 percent of respondents said they use technology primarily to detect claims fraud, a significant increase from 2016 and about half said they use it to combat underwriting fraud, up from 27 percent in 2016. (SAS)

10

11 of 40

Current Fraud Detection

Currently run your own Special Investigation Unit
The sole job of the unit is to monitor activity of suspicious claims

Use assistance from government officials as well as the National Insurance Crime Bureau to decipher fraud

Each employee working with claims gets trained to detect for fraud

Most employees in the unit are former law enforcement officials

11

Businesswire

12 of 40

Current Fraud Detection

10 percent or more of casual and property claims are fraudulent

Increase insurance rates, causes higher taxes and prices for consumers

12

Businesswire

13 of 40

Utilizing AI Technology to Prevent Insurance Fraud

13

According to Verisk analytics, when incorporating AI technologies into claims fraud analytics, it’s important to understand the four critical components of operationalizing them: variety, value, volume, and velocity

Value:

Nearly 60% of organizations suffer from inaccurate business intelligence because of slow or poor access to the right data

Variety:

System can automatically recognize suspicious patterns and characteristics indicative of fraud

Verisk Analytics

14 of 40

AI Fraud Components

14

Volume:

Analyzing the depth of information offers a more complete picture of a claim

Velocity:

Need the proper technology to enable the modeling and analyzing processes

Cloud Computing

Verisk Analytics

15 of 40

The Use of AI Fraud Prevention in Competitors & How it Works

Insurance Competitor CNA Financial

Uses Shift Technology to find the most suspicious claims with predetermined analytical paths.
It is a predictive analytics application
Became the first company to use Shift Technology’s FORCE software solution.
Improved customer experience and was a great addition to the improvement of CNA

15

CNA Financial

16 of 40

Shift Technology

16

Shift Technology

Generates a fraud score for each claim that indicates how suspicious the claim is, contributing factors, and how the claim could be investigated.

Shift Technology

17 of 40

The Utilization of Shift Technology

17

Shift Technology

18 of 40

18

Our Purpose

Evaluation

Our Solutions

Key Findings

Evaluation

Next Steps

Conclusion

19 of 40

Pros Vs Cons

19

Cons:

Unemployment
75% Hit rate

75% of claims Shift flags are fraudulent

Pros:

Decrease Fraud
Pays For Itself
More Happy Customers

US Bureau of Labor Statistics

Camille

Cons:

Unemployment

Shift technology certainly minimizes error and detects fraud in a drastically quicker amount of time than an employee at your company typically could. However, the implementation of this AI software eliminates the need for an employee to manually investigate claims. According to the US Bureau of Labor Statistics, the unemployment rate in America rose up to 14.9% in April due to the pandemic. Our economy is still in recovery, on a decline to 6.9%. This raises the question, is it smart to implement technology that could eliminate the jobs of more Americans?
https://www.bls.gov/charts/employment-situation/civilian-unemployment-rate.htm

75% hit rate

According to Shift, 75% of the claims Shift flags are identified as fraudulent. That poses the question, are employees wasting time that they could use to devote on more important tasks? Next I am going to give a counterargument to these points

Pros:

Despite the Cons that I just highlighted, there are major pros that we believe out weigh each con. First and foremost, the biggest pro that arises from the implementation of Shift is the fact that fraud will take a dramatic decrease. It will not only be easier to identify but it will improve efficiency, saving time for agents to do other things such as tend to the relationship with current customers as well as gaining new ones.
According to the FBI insurance companies lose over 40 billion dollars in fraudulent claims. Since your company is the second largest player when it comes to insurance in the US, it is safe to say that a large portion of this over $40 billion dollars is incurred by your company. If shift is implemented, then the yearly savings will be far greater than the initial price of implementation.
Lastly, you will be able to increase the amount of happy customers. This will happen in two ways, the amount of saving that your company will be able to have by decreasing just a fraction of the money spent on Fraud will allow them to decrease the cost of premiums for their current customers by anywhere between $400 - $700. Therefore, current customers will stay and brand loyalty will grow all while they attract new customers because of their competitive rates.

20 of 40

Final Recommendation

We recommend Geico to implement Shift’s FORCE technology

20

FORCE Solution Sheet

21 of 40

Final Recommendation

The FORCE technology will:

Leverage the power of internal data
Delivers actionable claims alert via an intuitive, customizable user interface
Help Geico stay ahead of emerging fraud scenarios
Improve investigative efficiency
Lower false positives
Detects fraud with twice the accuracy of other solutions

How it will contribute to a competitive edge for Geico:

Geico’s other competitors are not utilizing this technology which makes it inimitable
Geico’s services and products will be more valuable with the utilization of this technology
There will be a quick return on investment (about 4 months to be exact)
The improved efficiency and accuracy will add value to Geico as a company

21

FORCE Solution Sheet

22 of 40

22

Next Steps

Our Purpose

Our Solutions

Our Process

Evaluation

Next Steps

Conclusion

23 of 40

Implementation of FORCE

23

2. Cost Benefit Analysis

3. Financing Our Solution

1. Onboarding

24 of 40

Onboarding & Integration

Simple Integration

FORCE is a Web-Based, Cloud SaaS solution. This minimizes IT involvement and costs for GEICO at both the initial implementation and for maintenance.

Efficient User-Interface

Helps GEICO’s SIU handlers by automatically identifying possible fraud cases, explaining why the claim was flagged, and recommending “next steps” for investigating the claim.

Quick Onboarding

GEICO simply needs to provide its historical policy and claims data in any format and FORCE ingests, cleans, and maps the data in as few as four months.

24

Claims Fraud in the Digital Age

FORCE Solution Sheet

25 of 40

Cost Benefit Analysis

FORCE’s solution is Financial Model Based - GEICO would pay on a per claims basis, and have the ability to end the service any time without penalties

COSTS

BENEFITS

25

Digital Insurance Agenda

Claims Fraud In the Digital Age

QUICK RETURN ON INVESTMENT:

In 2016, a leading insurance provider, similar to GEICO purchased FORCE

Within four months of going live FORCE delivered a positive ROI
Had a yearly ROI of 500 to 600 percent

PREVENT REVENUE LEAKAGES:

Globally, insurers lose hundreds of billions in revenue due to fraudulent claims
Fraudulent claims total at least $40 billion per year in the United States

FORCE Solution Sheet

26 of 40

Financing Our Solution

26

Observations:

GEICO increases advertising expenses year over year
The increase in advertising does not reflect the increase in premiums

Result:

Easily Divert funds from advertising to investing in Shift’s FORCE solution

(Mohamed)

We decided that when it came to analyzing GEICO’s finanances, we had to touch on the large amount of expenses spent on advertising, and we thought it would be best compared to their top competitor, Prograssive
Both GEICO and Progressive increase their advertising expenses year over year
The increase in advertising is both for getting new members and maintaining current members
However studies show that in more recent times, the advertisements keep customers more than they attract new customers.
to begin talking about our Next Steps

Thank you very much Amanda. Now I will go into how we recommend you finance this initiative moving forward. These two graphs represent the amount of money spent on advertising in comparison to premiums written. On the left we see your spending on advertising is increasing year over year, yet the percent of premiums written is not increasing. We decided to compare your advertising costs to your main competitor, Progressive. Progressive has a similar story where spending on advertising year over year increases but premiums written begin to decrease in the last three years.

Therefore we recommend to divert $100 million from the advertising budget to this initiative because there is not a direct correlation between money spent on advertising and premiums written.

Now I will pass it off to Camille to conclude.

Make recommendations more specific
Get values for how much you want to take away
Make charts bigger

27 of 40

27

Conclusion

Our Purpose

Our Solutions

Key Findings

Evaluation

Next Steps

Conclusion

28 of 40

Final Overview

28

PROBLEM STATEMENT

How can GEICO use technology or product innovation to gain a competitive advantage in the insurance industry. What are viable options for increased efficiency, or long-term sustainability?

FINAL SOLUTION:

Invest in Shift’s FORCE solution

Automated, AI-native & SaaS-based fraud detection solution

Why?

Reduce the hundreds of billions revenue losses that stem from fraud
Quick Return on Investment
Gain a competitive advantage

FORCE Solution Sheet

Camille

Finally, to wrap up our presentation, we examined a problem your company faces in regards to technology, product innovation and gaining competitive advantage in that area to excel in the insurance industry. Shift’s technology, in a nutshell, would involve integrating automation into your current fraud prevention process. 10% or more of claims are currently fraudulent, which is a clear problem in the insurance industry and why our team recommends that you implement this technology. Shift’s technology would first allow your company to decrease manual labor involved with identifying these claims with their detailed AI technology that flags claims instantaneously. Shift’s FORCE solution would reduce the hundreds of billions of losses in revenue that stem from fraud, have a quick return on investment, and allow your company to gain competitive advantage in the insurance industry.

https://www.shift-technology.com/

https://2cdn6p3duuk84f0po2489ytp-wpengine.netdna-ssl.com/wp-content/uploads/2019/09/Shift_Force_2Pager_EN_RGB-1.pdf

29 of 40

29

Any Questions?

Thank you!

30 of 40

Citations

Digital disruption in insurance: Cutting through the noise. (n.d.). Retrieved October 18, 2020, from https://www.mckinsey.com/~/media/mckinsey/industries/financial%20services/our%20insights/time%20for%20insurance%20companies%20to%20face%20digital%20reality/digital-disruption-in-insurance.ashx

GEICO At A Glance. (n.d.). Retrieved October 19, 2020, from https://www.geico.com/about/corporate/at-a-glance/

INNOVATION IN INSURANCE: HOW TECHNOLOGY IS CHANGING THE INDUSTRY. (n.d.). Retrieved October 18, 2020, from https://www.iif.com/portals/0/Files/private/32370132_insurance_innovation_report_2016.pdf

The Insurance Savings You Expect. (n.d.). Retrieved October 19, 2020, from https://www.geico.com/

Property, Casualty, and Direct Insurance in the US. (n.d.). Retrieved October 19, 2020, from https://my-ibisworld-com.proxy-um.researchport.umd.edu/us/en/industry/52412/about

Tzanetos, G. (2020, August 28). Geico's Marketing Strategy Through Advertisements. Retrieved October 19, 2020, from https://www.investopedia.com/articles/insurance/020117/look-geicos-marketing-strategy.asp

Walgrove, A. (2020, August 13). Importance of Brand Awareness: Advantages and Perks: Taboola. Retrieved October 20, 2020, from https://blog.taboola.com/importance-of-brand-awareness/

https://www.spglobal.com/marketintelligence/en/news-insights/latest-news-headlines/ad-spending-at-state-farm-progressive-tops-1b-in-2019-geico-nearly-hits-2b-57549297

30

31 of 40

References

31

32 of 40

32

Appendices

33 of 40

Appendix 1:

33

34 of 40

Appendix 2:

34

35 of 40

Appendix 3:

“On a test set of more than 1M documents, the AI module had a prediction rate of 81% with an accuracy of 97%, even higher than human claim handlers. For a volume of 1M claims per year, this saves a workload worth of $50M and enables to reduce leakage by more than $20M per year.” - Digital Insurance Agenda

Helps explain why GEICO will quickly receive their ROI when they invest in Shift’s FORCE solution

35

36 of 40

AI in Social Media

36

Regular software is coded by humans and can only be improved manually by humans.
AI technology and tools, “can improve on their own, based on both their own historical performance and new data given to the system—unlocking potentially unlimited performance gains.” (Marketing Intelligence Institute)
Geico has to take advantage of the way that AI technology works and expand it into their social media marketing.

Unimetric

Geico heavily relies on their marketing strategy to bring new customers in. They are one of the only companies that spends over a billion dollars on advertising annually to make sure customers understand the services they provide. However, Geico lacks on their overall social media presence.
Bringing AI technology into Geico’s social media will create a lasting impact on the company. That means every piece of marketing software you use today, from ad buying to analytics to automation to content strategy to social, can be made more intelligent using AI and machine learning.
And, across all social media platforms, AI and machine learning are regulating how the content you create and the ads you buy are placed in front of users—often in ways that aren’t entirely transparent to marketers.
As seen in the image Geico is at almost the lowest regarding their twitter presence.
Relating to the audience that is on twitter and on social media platforms could create a huge boost for geico if they took advantage of the technology.
Insurance isn’t an attractive or generally engaging market. Geico has used TV Ads and the gecko to create entertainment around their product but there is still more they can do.
I will now pass it onto caroline to further discuss how AI can be used in social media apps

37 of 40

Marketing Solution

Geico has used 3 main strategies in their marketing strategies:

Geico has relied on a simple, always relevant brand promise of ease and savings.
They have remained committed to their simple message.
Geico creates funny, clever content that makes the insurance industry seem less boring and more intriguing (Hawk Partners).

The Geico advertising campaign strategy is to saturate the market with print advertising, radio ads, and television advertisements (IP Watchdog).
Solution: Geico should implement AI technology with their social media.

With AI technology, Geico can utilize it to help with content generation, trend finding, ad targeting, PPC ad generation, better understanding their consumer market, and boosting return on investment.
AI technology will help with leveraging the customer experience in marketing and will be using big data analytics, machine learning, and other processes to gain insight into Geico’s target audience (Forbes).

37

Thanks Julia.
Fifteen minutes could save you 15% or more on car insurance
Geico has leveraged one of the longest-running taglines and call-to-action strategies in history, resulting in a recall rate of 90%
but Geico uses these lovable characters perfectly to provide a funny anecdote that brings the brand promise to life – which is what keeps people rushing to YouTube to watch these characters, again and again.
They have some more serious ads aimed at older drivers, and others that are the more humorous and/or unexpected ad campaigns that are designed for younger drivers. However, with social media advertising becoming more and more present in our lives, especially with targeting the main generation, Geico must begin to work on their social media advertising.
Geico could use digital marketing AI to capitalize on a marketing segment they’re lacking in
They have great ads, but their digital marketing is lacking

Digital marketing is on the rise and becoming extremely important for companies
Their brand voice and image is inconsistent, and frankly, non-existent
On one instagram post, they only got 267 likes. With a following of 38.8k, that is about a .6% engagement rate.

Through utilizing AI technology w/ social media as well as putting a greater focus on social media advertising, Geico can use their iconic humor and brand voice to engage consumers that they don’t normally, such as younger people

38 of 40

Claims Management Solution: H20

Implement AI in Claims Management through H20 Platform
Driverless AI Software

Automates data science and machine learning workflows
Machine learning interpretability

Benefits for Geico

Reduces risk of fraud
Streamlines claim process
Makes assessment and payment of claims for customers fast and accurate

38

Thanks Mohamed. I’m now going to introduce our solution that incorporates AI into claims management. Our solution is to incorporate the H20 platform into Geico’s claims management process, which is an AI machine learning software known as Driverless AI. In fact, Progressive already utilizes this platform to improve the claims process. Driverless AI, according to H20, automates difficult data science and machine learning workflows and also offers machine learning interpretability. Essentially, for the claims management process, the AI software would be able to interpret data input from customer claims and provide the following benefits to Geico.

First, the software reduces the risk of fraud. The AI scores claims for issues such as fraud where the claims with the lower probability of an issue are processed automatically, and the claims with a higher probability of fraud are routed to investigators.

Second, the software streamlines the claims process for Geico. Traditional claims management is completed manually which can slow down the process. AI software that streamlines this process allows Geico employees to focus on more important tasks such as client satisfaction, or focus on claims flagged by the software.

Last, this solution relates back to Geico’s relationship with customers. The way the claims management process is handled is essential for retaining customers and keeping them satisfied. The Driverless AI software ensure the assessment and payment of claims for customers is fast and accurate.

39 of 40

AI in Claims Management

Increase Efficiencies
Increase Customer Satisfaction
Decrease Fraudulent Claims
Improve Metrics

39

Fraud

First things first, I would like to explain to you how AI can work when it comes to claim management. In order to do so, I will walk you through the image to your right. Ai takes in the real time claims data, analysis it, creating the specific graphs that you seek to find, classifies the claims, and then is used in a multitude of operations.
When it comes to how Geico utilizes their agents we can look at the time each agent allocates to claim management. This part of their job takes up a significant amount of their time while not taking up a significant amount of brain power. If Geico can utilize artificial intelligence to help streamline the claims process then the agents can spend all the time and energy devoted to the management on these claims elsewhere. Ultimately this will allow agents to perform their other tasks such as comforting clients, gaining new clients, and maintaining their current client list. This allows me to move into the second aspect.
With a streamlined claims process, clients will seamlessly file their claims, have it reviewed and the entire process will take a lot less time and energy. By eliminating the stress that comes with dealing with agents after an accident, customers can prioritize their focus on their wellbeing as well as their vehicle. The ease of the process is sure to increase customer satisfaction but another aspect AI can help is decreasing the amount of fraudulent claims.
When it comes to fraudulent claims AI can help medicate the risk and of situations like that occurring by using analytics to calculate the likelihood of a fraudulent claim and then an agent can get involved if a claim has a high risk of being fraudulent. Again this will play back to the last two points I brought up. Efficiency will increase and agents will be able to prioritize difficult cases or ones where fraud is likely.
AI will be able to not only benefit in the ways listed but this last point is an extremely important one. When AI is integrated into the claims process, the data that AI can use will allow the AI to improve metrics all over the charts. The optimal monthly qute, what should deductibles be priced at, how to maintain current customers, how to attract new ones, etc. The ability that AI can have when it comes to analyzing the metrics around claim management are real and significant.

Notes: Robotic Process Automation (RPA) can empower insurers in building a high-growth responsive business while optimizing cost.

whereby they guarantee payment in case of damage or financial loss and accept the financial risk for liability arising from such guarantee

40 of 40

Utilizing AI Technology to Prevent Insurance Fraud

40

When incorporating AI technologies into claims fraud analytics, it’s important to understand the four critical components of operationalizing them: variety, value, volume, and velocity

Variety:

System can automatically recognize suspicious patterns and characteristics indicative of fraud
Takes a lot of time and energy
Types of variety:

unstructured data, such as images, audio, video, text, and social media

Value:

Need correct insight from claims fraud AI
Nearly 60% of organizations suffer from inaccurate business intelligence because of slow or poor access to the right data
35% of large property/casualty insurers and 24% of midsize carriers are expanding their data science efforts.

Volume:

Need deep enough data
Claims often include depth of information
Analyze the depth of information offers a more complete picture of a claim

Velocity:

Processing speed is vital
Need the proper technology to enable the modeling and analyzing processes

Cloud Computing

Real time insights