Santam Insurance boosts customer service and beats fraud

Using IBM SPSS predictive analytics to identify risks and accelerate claims settlement

Published on 07-Jul-2011

"In the first month of using the SPSS solution, we were able to identify patterns that enabled us to foil a major motor insurance fraud syndicate. Within the first four months, we had saved R17 million on fraudulent claims, and R32 million in total repudiations – so the solution delivered a full return on investment almost instantly!" - Anesh Govender, Head of Finance, Reporting and Salvage, Santam Insurance

Customer:
Santam Insurance

Industry:
Insurance

Deployment country:
South Africa

Solution:
BA - Business Intelligence, Big Data & Analytics, Big Data & Analytics: Operations/Fraud/Threats, Information Integration, Smarter Planet, Optimizing IT

Smarter Planet:
Smarter Insurance

IBM Business Partner:
Olrac SPSolutions

Overview

Founded in 1918, Santam has grown to become South Africa’s largest short-term insurance company. With more than 650,000 policy holders and assets under management of 17 billion South African Rand (US $2.4 billion), the company enjoys a market share of more than 22 percent. It offers customers a wide range of services in personal, commercial, agricultural and specialist insurance and risk management.

Business need:
Santam wanted to find a way to improve its service to customers by settling claims faster and keeping premiums low. To achieve this, the company needed to maximise operational efficiency and find smarter ways to combat fraud.

Solution:
Santam worked with Olrac SPSolutions, an IBM Business Partner, to design a claims segmentation solution based on IBM SPSS predictive analytics software. Each claim is automatically scored according to its risk level, and then distributed to the appropriate processing channel for settlement or further investigation.

Results:
Enhances Santam’s ability to detect fraud, foiling a major crime syndicate and saving 17 million South African Rand (US $2.4 million) in the first four months of operation.

Benefits:
Improves customer service by enabling legitimate claims to be settled within an hour, more than 70 times faster than before. Reduces the need for claims adjusters to visit clients to assess low-risk claims, significantly reducing operational costs.

Case Study

Founded in 1918, Santam has grown to become South Africa’s largest short-term insurance company. With more than 650,000 policy holders and assets under management of 17 billion South African Rand (US $2.4 billion), the company enjoys a market share of more than 22 percent. It offers customers a wide range of services in personal, commercial, agricultural and specialist insurance and risk management.

Operating in a market where fraudulent activity can account for an estimated six to ten percent of all premiums, Santam’s claims division faced a considerable challenge.

“On the one hand, we need to assess claims carefully to avoid exposure to fraudsters, which affects our bottom line and increases the cost of premiums for our good customers,” explains Anesh Govender, Head of Finance, Reporting and Salvage at Santam. “On the other hand, to maintain our excellent reputation for customer service, we need to be able to settle legitimate claims quickly. We wanted to find a way to treat each case on its merits and deliver a better service in terms of both speed and safety.”

A new operating model
The claims division began to develop a new operating model, which would pass claims down different channels depending on their assessed level of risk. The five channels, ranked from lowest to highest risk, are:

  • Immediate: these claims are settled as quickly as possible, without further assessment.
  • Digital: these claims require photographic evidence, which is analysed and verified by a team at Santam’s head office.
  • Mobile: these claims are assessed by a Santam operative who visits the site of the claim.
  • Complex: these claims require a more sophisticated assessment by an expert team, usually due to legal issues.
  • Merit: these are the claims that are considered most likely to be fraudulent, and are passed to a special unit for thorough investigation.

Leveraging predictive analytics
To ensure that claims would be handled by the appropriate channels, Santam needed to find a method of segmenting claims as quickly and effectively as possible. The company became interested in the possibility of using predictive analytics software to manage this segmentation. Following a detailed evaluation of the analytics software available on the market, the Santam team drew up a shortlist of two vendors: SAS, and Olrac SPSolutions (formerly known as SPSS South Africa), who proposed a solution based on IBM SPSS software.

“We ultimately chose Olrac SPSolutions for two main reasons,” explains Anesh Govender. “First, we believed that IBM SPSS offered a better range of functionality, particularly in terms of its ability to integrate with our core claims management application, which runs on a mainframe platform. Second, we were hugely impressed with the team from Olrac SPSolutions. Even though we were proposing to build a very innovative and leading-edge solution, their technical skill and business experience gave us a lot of confidence that IBM SPSS could deliver what we needed.”

Proving the concept
To make a solid business case for the adoption of the IBM SPSS solution, Santam’s in-house team worked with Olrac SPSolutions to deliver a proof of concept that focused on one key business area: personal motor insurance for accidental damage, collision and overturning.

“We were able to demonstrate the power of using predictive modelling and business rules to score claims based on a number of known risk factors,” explains Anesh Govender. “For example, based on existing claims data, we already knew that statistically most car accidents occur around 10am, but that most fraudulent ‘accidents’ happen between 10pm and 5am. So the time that the accident is reported is a key factor in determining the risk score of the claim. By building business rules based on numerous risk factors like this one, we were able to develop a reliable model that enabled us to segment claims effectively.”

Easy expansion
Following the proof of concept stage, Santam worked with Olrac SPSolutions to move the solution into the production environment, where it would handle claims segmentation in real time. After a successful go-live, the company began extending the solution to handle the whole of its motor insurance business, as well as other business lines.

“The initial project took six months, because it required the development of some quite complex interfaces with our mainframe,” comments Anesh Govender. “We were then able to re-use these interfaces in the next project, so it only took two months. Olrac SPSolutions helped us lay the groundwork very efficiently, so from now on, each new project should be quicker and easier than the last.”

Increasing flexibility
Santam is currently in the process of upgrading the solution to IBM SPSS Decision Management 6, which will offer further flexibility, especially in terms of creating and managing the business rules that govern the claims processing channels.

“At the moment, some of the workflows are still driven by mainframe systems, which makes it relatively difficult to adapt them to changing business needs,” says Anesh Govender. “With SPSS Decision Management 6, we will be able to manage almost the entire process within SPSS itself, which will give us much greater agility and flexibility.”

Faster service, huge savings
The current version of the SPSS solution has already delivered spectacular real-world results for Santam in terms of fraud detection, customer service and return on investment.

“Before this solution, the minimum time it took to settle a claim was three days,” explains Anesh Govender. “Now, the low-risk claims that pass down the ‘immediate’ channel can be settled within an hour – so customers with legitimate claims get much faster service. This has also allowed us to significantly reduce the number of claims that need to be assessed by mobile operatives, which will lead to considerable operational cost savings.

“The most startling results, though, have come as a result of enhanced fraud detection. In the first month of using the SPSS solution, we were able to identify patterns that enabled us to foil a major motor insurance fraud syndicate. Within the first four months, we had saved R17 million on fraudulent claims, and R32 million in total repudiations – so the solution delivered a full return on investment almost instantly!”

Looking to the future
Over the next two to three years, Santam plans to enhance the claims segmentation process even further by using the sophisticated predictive modelling capabilities of SPSS to fine-tune the scoring process.

“At the moment, the scoring process is based primarily on sets of predefined business rules – but as we process more claims and get more data in the system, we will be able to build more complex predictive models,” says Anesh Govender. “We’re particularly interested in building propensity models to look at customer behaviour. This would enable us to group customers into different risk profiles, so that we can segment our best customers and offer them more tailored products and higher service levels. We may also be able to build models using seasonal data and weather data to help us react more effectively to catastrophes such as fires and floods.”

He concludes: “IBM and Olrac SPSolutions have helped us build a solution that has not only transformed our claims processing methodology in terms of speed and efficiency, but also provides game-changing insight – helping us identify suspicious claims more quickly and protect our business and our customers by bringing fraudsters to justice.”

Smarter Insurance: Fighting fraud and boosting customer service
Instrumented: When a claim is submitted, Santam captures data related to a number of key risk indicators and automatically transfers it to the new SPSS solution.
Interconnected: The analytical engine uses a combination of business rules and sophisticated predictive models to assess claims for potential fraud and transfer them to the appropriate processing channel.
Intelligent: By segmenting claims according to risk factors, Santam can focus on investigating high-risk claims and catching fraudsters, while rewarding good customers with fast settlement and better service.

Products and services used

IBM products and services that were used in this case study.

Software:
SPSS Decision Management

Legal Information

© Copyright IBM Corporation 2011 IBM Southern & Central Africa Private Bag x9907 Sandton South Africa 2146 Produced in South Africa July 2011 All Rights Reserved IBM, the IBM logo and ibm.com are trademarks of International Business Machines Corporation, registered in many jurisdictions worldwide. A current list of other IBM trademarks is available on the Web at “Copyright and trademark information” at: ibm.com/legal/copytrade.shtml. SPSS is a businessmark of SPSS, Inc., an IBM Company, registered in many jurisdictions worldwide. Other company, product or service names may be trademarks, or service marks of others. References in this publication to IBM products, programs or services do not imply that IBM intends to make these available in all countries in which IBM operates. Any reference to an IBM product, program or service is not intended to imply that only IBM’s product, program or service may be used. Any functionally equivalent product, program or service may be used instead. All customer examples cited represent how some customers have used IBM products and the results they may have achieved. Actual environmental costs and performance characteristics will vary depending on individual customer configurations and conditions. IBM hardware products are manufactured from new parts, or new and used parts. In some cases, the hardware product may not be new and may have been previously installed. Regardless, IBM warranty terms apply. This publication is for general guidance only. Photographs may show design models.