The steps to using analytics…successfully

Advances in technology, evolution of the distribution channels, demographic shift, economic conditions and regulations changes. How does an insurer prioritize all these seemingly competing goals and create sustainable competitive advantage. One answer is analytics.

Many insurance companies are just beginning to take steps toward becoming an “analytic insurer” – one that embeds analytics into daily operations to make better decisions that reduce costs, improve pricing, and more. And those organizations with more advanced analytic capabilities are actively seeking to build on previous successes and grow their analytic capabilities. And no wonder, given the increasing volumes of data being produced through enterprise business systems, online interactions, social media, and other channels.

Analytical Lifecycle

Implementing analytics is not as straight forward as it sounds. There are many steps in the analytical life cycle to consider, but essentially it can be broken down into four main sections:

1. Data preparation

2. Analysis and predictive modeling

3. Deployment

4. Model management


In a series of four blog articles over the coming weeks each of these areas will be discussed in more detail.

Turning the increasing volumes of data into useful information is a challenge for most organizations, but following these simple steps insurance companies will be able to implement analytics…successfully.

I’m Stuart Rose, Global Insurance Marketing Director at SAS. For further discussions, connect with me on LinkedIn and Twitter.

Post a Comment

Data governance - the new prodigal child

The old adage is that “Data is the lifeblood of the insurance industry.” However, for many insurance companies, data is like the red-headed stepchild. No one is willing to take care or have responsibility for it. In the past, insurance companies have created data governance programs, but these have often failed or underperformed due to either:

  • Business executives and managers consider data to be an “IT issue.”
  • The return on investment (ROI) for data governance isn’t clear.
  • Linking governance activities to business value is difficult.
  • Organization structures are fragmented with multiple data silos.

data governance


Fortunately, today, things are changing.   C-level executives recognize the need to manage data as a corporate asset. In fact,  you will see a Chief Data Officer (CDO) appearing in boardrooms of many organizations. The primary responsibility  of the CDO is to ensure that data will be managed as a shared asset to maximize business value and reduce risk.



To help achieve this objective, the CDO will create a team of data stewards who are go-to data experts serving at the point of contact for data definitions, usability, questions and access requests. A good data steward will focus on:

  • Creating clear and unambiguous definitions of data
  • Defining a range of acceptable values, such as data types and length
  • Monitoring data quality and starting root cause investigation when problems arise.
  • Understanding the usage of data in the business units.
  • Reporting metrics and issues to the data governance council.

As insurance companies continue to become more data-driven, their success will ultimately hinge on the ability to maintain and use a coherent view of this data. Better data can drive insight and help insurers make better decisions. To learn more about on this subject download the white paper “SAS Data Governance Framework: A Blueprint for Success.”

Today, data governance is no longer an afterthought. With the emergence of Chief Data Officers, insurance companies are now treating data like the prodigal child.

I’m Stuart Rose, Global Insurance Marketing Director at SAS. For further discussions, connect with me on LinkedIn and Twitter.


Post a Comment

Insurance and the rise of the Chief Risk Officer

The role of insurance is to bring some predictability, manageability and stability to a chaotic and uncertain world.  In essence, it is a risk mitigation tool.

The role of the Chief Risk Officer (CRO) is to manage the overall risk strategy for the insurance company. They are responsible for defining the medium to long-term risk strategy for their insurance organization.

This strategy takes into consideration variables such as risk appetite, target market, customer segments, core products, distribution channels and expected return on investments. This is achieved by what is commonly referred to as the capital management and planning process.

The first step in the process requires insurers to identify and model all material risks that can potentially affect their solvency or the long-term value of equity. To have an efficient capital management framework, insurers also need to coordinate the actions of their risk units with their actuarial and finance departments. Planning and budgeting exercises that steer direction for operational actions should be coordinated with a view into risks, profitability and shareholder returns.

The second step is the necessity to align their decision-making process with estimates for how much capital the organization must have on hand in light of commitments and identified risks. This helps business line managers perceive the constraints and opportunities that economic capital presents in the areas of risk-based pricing, customer profitability analysis, customer segmentation and portfolio optimization.

With an effective capital management, insurers should be able to weather extreme internal risk events (e.g., a large operational risk event) and external scenarios (e.g., a catastrophic natural disaster) at an enterprise level. It also helps business line managers create favorable opportunities, as they can generate an optimized risk-return profile of their product portfolios.

ORSA dashboardA new responsibility for many CROs is the emerging Own Risk Solvency Assessment (ORSA) that is required for Solvency II and other insurance regulations. One of the fundamental requirements of ORSA is that companies conduct an annual, forward-looking assessment. The goal is not only to demonstrate that the company’s current capital needs are appropriate, but also that its future capital needs will be met over a specified assessment time frame (usually three to five years). The report also allows regulators to get an enhanced view of an insurer’s ability to withstand financial stress.

With the recently launched SAS Capital Planning and Management solution, CROs can perform the quantitative aspects of an ORSA, taking into account projected balance sheets, income statements and risk appetite with the capability for iterative scenario analysis and stress testing.

To learn more about emerging insurance regulations,  download the white paper “ORSA: The New Kid in Town”.”

Insurance and risk have always gone together.  However,  as insurance becomes more complex and sophisticated,  the rise of the CRO is inevitable. and the importance is undeniable.

I’m Stuart Rose, Global Insurance Marketing Director at SAS. For further discussions, connect with me on LinkedIn and Twitter.

Post a Comment

Welcome to “data-driven decisions”

Business analytics is about dramatically improving the way an organization makes decisions, conducts business and successfully competes in the marketplace. At the heart of business analytics is data.  Historically, the philosophy of many insurers has been on collecting data, data and more data. However, even with all this data, many insurance carriers are struggling to achieve better analytics and gain business insight simply because many insurers’ systems typically aren’t designed to master the data in such a way to make it useful for analytical purposes.

data driven


The best way to maximize the efficiency and effectiveness of data-driven decision making is to focus on determining the sufficient amount and quality of data necessary for satisfying the execution of a business decision. Of course, this decision-making strategy is easier in theory than it is in practice.


Data driven decision management comprises three components:

  1. Data requirements - Different decisions will have different data requirements, which include the data volume, variety and velocity necessary for data-driven decision making. Not every decision requires the same amount and quality of data.
  2. Decision criteria - For example, a decision that must be made within 30 seconds has very different data requirements than a decision that should be made within 30 minutes, or a decision that could be made within 30 days.
  3. Decision evaluation - The quality of a decision is determined by the business results it produces, not the person who made the decision, the quality of the data used to support the decision, or even the decision-making technique.

The big data movement has brought with it a host of new technologies and analytical capabilities. However,  these technologies are ineffective without the right questions and talent to create a data-driven culture.

Imagine…What if you could . . .

. . . predict the buying behavior and decision criteria of your customers and prospects weeks before your competition

. . . gain market share by storing and analyzing the explosion of telematics data for improved risk assessment and distinct competitive advantage

. . . price more accurately based on risk attributes, key demographics, competitors rates, demand elasticity models and make adjustments in real-time

It is possible.

To learn how to turn this theory into reality, download the white paper “Return on Information: The new ROI.”

I’m Stuart Rose, Global Insurance Marketing Director at SAS. For further discussions, connect with me on LinkedIn and Twitter.


Post a Comment

Helping small businesses through the insurance minefield

Insurance can be a complex business, so filing an insurance claim can be daunting task for many small businesses. When an incident does occur, be it property damage, business interruption, professional indemnity or public liability among the myriad of other potential causes of loss, it is typically a period of concern and uncertainty.

Here the hand-holding that brokers offer their clients can make a big difference, offering reassurance and advice. But it is a speedy settlement that is critical to getting a company back on its feet. Such an outcome is particularly important during the current economic environment, with banks continuing reluctance to lend to small and medium-sized enterprises (SMEs). A protracted claims process could prevent a small business from trading, threatening its survival.

The economic downturn has also had an impact on the risk profiles of many SMEs, with companies diversifying their offerings and becoming leaner. Some common examples include extending opening hours, plumbers turning their hand to fitting solar panels, bookshop owners serving hot drinks or newsagents bringing in cash or lottery machines. These changes demonstrate the spirit of SME innovation, however there are liability implications.

By maintaining a close dialogue with its broker channel, insurers can stay on top of these changes, ensure their clients have the right depth and breadth of coverage and therefore when losses occur, there is no cause for dispute or repudiation.

Claims “triage” or decision-tree analysis allows claims departments to determine which cases can be approved speedily while isolating those that require further investigation. Using key data, such as the size of a claim, past history etc enables claims handlers to choose the correct handling process. Such an approach can help avoid backlogs, reducing the claims lifecycle and ensuring a satisfactory and speedy settlement for small business customers.

Data visualization

SAS works closely with its insurance customers to apply analytics across the entire claims process, reducing loss ratios and lowering loss-adjustment expenses through the more strategic and efficient use of data. You can download the white paper Predictive Claims Processing and find out more at

Post a Comment

Is the customer experience overrated?

According to analyst firms, consulting companies and various other research, customer experience is the primary priority for insurance companies.  But is customer experience overrated?

Let’s start by considering the primary interactions between an insurance company and its customers: new business, billing, renewals and claims. Ask any insurance executive, especially property & casualty, and they will tell you the most important customer experience is the claims process. Yet a recent survey by Accenture found that more than 40% of customers who submit claims are likely to switch insurers within the following year, regardless of satisfaction.

If the claims satisfaction does not positively influence retention rates, we should ask if the customer experience is just hype?

In an article, “Does Improving the Customer Experience Matter?”, Mark Breading from Strategy Meets Action, writes that to answer this question insurance companies must consider three questions:

  1. Who is the customer?
  2. What influences their experience?
  3. Why does it all matter?

It’s the third question I want to focus on. In a blog I wrote earlier this year, I discussed that the primary objective of an Insurance CEO is to grow the business. For most insurance companies, this means increasing market share. To attract new customers and retain existing policyholders, insurance companies need to differentiate themselves from their competitors and the customer experience is vital to achieving this objective.

To learn more download the white paper: “Six Steps to an Unmatched Customer Experience.”

As the great philosopher, Homer Simpson, once said “People can come up with statistics to prove anything”. So while the report from Accenture may be true, insurance companies know that the customer experience does indeed matter.

I’m Stuart Rose, Global Insurance Marketing Director at SAS. For further discussions, connect with me on LinkedIn and Twitter.


Post a Comment

Getting personal at the touch of a button

Over the past decade in insurance, the rise of the aggregators (organisations that compare quotes between different insurers) and direct channels has had a profound impact on personal lines distribution in the UK.

However, personal lines brokers remain a critical route to market, especially at a time when many insurers have been centralising their offerings and closing local branch offices. In terms of distribution, brokers remain responsible for placing 35 per cent of personal lines business, according to Datamonitor.

And increasingly, this business is placed online, via aggregator websites, insurer extranets and comparative quote systems. Personal lines brokers have embraced the full cycle EDI model (electronic data interchange; an electronic communication system that provides standards for exchanging data via any electronic means), yet the ability to maintain a close dialogue with insurer partners remains critically important.

Beyond offering competitive pricing, successful e-broking requires referrals to be handled efficiently while the ability to tap into quote enrichment methodologies also improves the journey. Processes around electronic trading and product delivery are constantly improving, including the ability to cross-check information electronically.

Within motor insurance, the Driver and Vehicle Licensing Authority (DVLA) MyLicence initiative was launched last summer. It allows brokers and insurers to access instant DVLA data on driving entitlements, motoring convictions and penalty points at the point of quote. The industry, through the Association of British Insurers (ABI) and the Motor Insurers’ Bureau (MIB) has worked jointly with the DVLA and Department of Transport to develop the data sharing programme.

360 customer view

Credit scoring and telematics are other examples of how big data can be used to validate application information and reduce fraud. Third-party data can help establish identity, honesty and level of risk at the touch of a button without having to rely on the customer to go through lengthy questionnaires.

By offering intermediaries the means to access better analytics, insurers have an important role to play in improving the profitability of their core broker accounts. And by maintaining a close dialogue, insurers will be in a better position to tailor their products, via schemes and affinities, to meet the changing needs of their brokers’ client base, allowing them to exploit opportunities as they arise.

SAS Visual Analytics works with insurers to offer more precise insights based on all available data. Our insurance software allows clients to take full advantage of big data for telematics, social media analysis, catastrophe modelling and risk analysis. Download our white paper ‘What Does Big Data Really Mean for Insurers?’ and find out more about Insurance Solutions from SAS.

Post a Comment

5 steps to fostering adoption of fraud analytics

Special Investigation Units (SIU) are extremely process-oriented and follow well-documented procedures to decide when a claim should be referred for investigation and what actions should be taken. Most of the staff are seasoned investigators who may be more inclined to trust their experience and tried-and-true processes than analytical techniques that may seem unintelligible.

They can’t trust the data if they don’t understand it - raw analytical model output is often full of technical jargon that’s not easily understood by the business user. This creates frustration for the end user - often an analyst or investigator with more business expertise than technical or statistical experience – and may lead them to think that the analyses are meaningless and analytics lack any real business value. To convince these users to adopt analytics, the solution must give them easily consumable information.

And, you must integrate the analysis with the business process. Fraud analytics solutions need to meet the investigators’ business requirements and be structured in a way that makes sense to them – not just to the data scientists and IT teams. They need a well-configured user interface (UI) that gives them all the information they need to make a decision about whether or not to proceed with an investigation.

Here are five things you can provide to help them see analytics as the answer:

  1. Provide a “one-stop-shopping experience.” The investigator should be able to complete the triage and review process without ever leaving the UI. In a manual fraud detection environment, the investigator has to access multiple internal and external data sources. By providing instant access and a holistic view of these data sources, you exponentially increase investigator efficiency and start to win converts.
  2. Remove the technical jargon. Use scorecards that speak the business language to show the investigator the rules or scenarios that were surfaced by the model and how much weight each contributes to the overall fraud propensity score. This will help them understand why a referral scored high for fraud risk and allow them to focus their investigation on the most critical factors.
  3. Show and tell. Analytical output surfaced in the scorecard should be supported by other data in the UI. Incorporate sections or links to reports that contain the data that support the rules and scenarios triggered in the scorecard. Investigators will not – and should not – blindly accept that the fraud scenarios surfaced in the scorecard are factual. Providing easy access to this data also helps facilitate their investigations.
  4. Leverage third-party data sources. Many SIU’s invest in expensive external data sources and most aren’t integrated with internal data sources. The lack of integration makes it difficult to get the most value from the data. Insurers can get more value from their investment by incorporating it into the UI, combining it with other data elements, and displaying it in a meaningful and user-friendly format. Common examples of this third-party data could include medical bill audit data, industry watch lists, sanction lists, loss history data, estimate data and public records information.
  5. Give them more than just a score. Provide deeper insight into suspicious activity by incorporating data visualization techniques such as link analysis, maps, graphs, charts, annotations, and highlighting of key text and other data elements that support the scorecard findings. These simple approaches resonate with the end user and add tremendous value by directing attention to the most relevant information.

The most powerful analytics in the world have little value if they can’t be readily understood and adopted by business users. Start with these five tips to get your investigators excited about fraud analytics. These will help them see that analytics can drive better decisions and better outcomes.

What do you think the biggest challenge is to the adoption of analytics? Tell us in the comments.

Post a Comment

Internet of Things: A game changer for insurance

Amol Kokane, SAS

Amol Kokane, SAS

~ This article is co-authored by Binod Jha, Global Product Manager for Insurance Solutions at SAS, and Amol Kokane, Senior Development Manager for Insurance & Risk Management Solutions at SAS ~

How might insurance policies change if sensor data could be automatically transmitted and analyzed from your car, your home and even your clothes?

The Internet of Things (IoT) has the potential to transform both the business and the IT infrastructure of the insurance industry. The data captured and transmitted by multiple sensors and devices – each attached to an associated, insured risk – will redefine existing processes of underwriting and pricing.

Let's look at a few definitions before we get into the possible affects of IoT on the industry.

Insurance premium pricing is based on a set of rating parameters that are derived from predictive models built from historical loss data. The law of large numbers and the associated spread of risk exposures helps in arriving at the pure premium for a pool of homogenous risks. But it’s very difficult to calculate loss propensity at each risk and exposure level in this pool.

Underwriting and pricing happens when a policy is sold, but the risk (both in quality and behaviors) changes over time. This fact should be (but seldom is) factored in the underwriting on a regular basis. So while the insurance industry has historically been data intensive, its ability to monitor claim propensity at each risk and exposure level during the lifetime of the policy is limited. As such, premiums are subsidized by good risks, taking the same hit by association with the bad risks. 

How does IoT change the game?

IoT makes it possible to capture rating parameters as well as the digital shadow (the historical data captured from devices and sensors) of insured objects and people, in real time. This huge amount of data can be analyzed for underwriting and pricing decisions at each exposure level, that is for each instance of the policy coverage period. This moves insurance to the realm of continuous calculations - thereby reducing incurred losses with proactive strategies vs. historical tactics. The IoT ecosystem has already matured enough to enable the early movers in this new IoT operating model for insurance. 

What are the benefits of IoT? 

For insurers, the benefits of analyzing IoT data include:

  • Lower claim severity and frequency. Continuous monitoring of insured risks, real time prediction of loss propensity, and reliable loss control mechanisms will impact claim severity and frequency.
  • More accurate risk assessment. Huge amount of granular data generated by sensors and devices attached to insured risks improves risk assessment accuracy and thus the underwriting process. In turn, this also reduces the suppression of material facts related to risk by customers at the time of policy acquisition.
  • Improved claim servicing. IoT enables automated loss notification in case of an accident or hazardous situation. The details related to loss or damage are captured by sensors and devices. Claim processing cycle time and loss adjustment expenses will be reduced as a result.
  • Higher customer satisfaction. Real time monitoring of insured risks and dynamic pricing creates transparency to both the underwriting and rating process. Educating customers on the factors that impact their specific insurance premium not only takes the required precautions needed to avoid premium loading, but also puts customers in more control of their plan. And it’s expected that with policyholders in the premium driver’s seat, so to speak, they’d be more satisfied, happier customers.

The entire insurance value chain – underwriting, policy servicing, claims, actuarial, reinsurance and customer service – are all impacted, maybe even disrupted, by IoT. Let’s focus how this may play out for a few lines of business.

Life and health insurance

Life and health insurance is expected to see the biggest impacts from IoT for a couple of reasons. Risk quality changes with lifestyle events (marital status) as well as with demographics (like age and occupation) impacts the baseline underwriting of the policy. With IoT, however, continuous monitoring of risk quality becomes a reality. The underwriting and pricing of insurance premiums won’t be limited to the inception of the policy – but will change during the entire coverage period. In this new world, the insurance premiums will fluctuate just as with monthly utility bills.

And the days may not be too far off when IoT wearable devices capture significant measures, like heartbeat, temperature, blood sugar, exercise duration and report them to insurers. Or when clinical diagnoses are be made from non-invasive methods, like analysis of sweat or tears. Premiums might even be calculated on a daily basis, transparent to the end customers - encouraging healthier lifestyles perhaps and reducing premiums accordingly.

Auto insurance

The first round of business model disruption has already taken place in the automotive side of insurance with the popular adoption of usage-based insurance (UBI) over traditional pricing in countries like the US, Canada, UK, Italy, Germany, and others. With UBI, vehicles are fitted with sensors that monitor driver behavior, to keep track of when, where and how the vehicle is in motion. The insurance premium is primarily determined on the basis of driving behavior, rather than proxy variables like vehicle make, model, and year. Enhancements to claims processing will likely be the next major improvement area for auto insurance – with automated notifications from these same telematics sensors for accident occurrence or non-typical driver profiling, helping carriers pinpoint events like accidents or theft. 

Property insurance

It’s expected that property insurance will see an increased adoption of IoT in this decade. Smart homes, offices, commercial buildings and industrial installations can be fitted with sensors and devices that generate real time data on hazards from overheating malfunctions to building material strength. Not only can any accidents or breakdowns be immediately recognized, but gradual deterioration can be monitored to avoid substantial damages - bringing insurers into proactive service for property customers.

Home, office and industrial equipment can all be remotely monitored for ongoing situational assessment. Any measured factor leading to malfunction or damage can trigger notifications to repair shops or manufacturers for prioritized service. Smarter carpets that detect falling incidents can notify medical assistance, reducing staged accidents and liability claims. The scope of pragmatic applications is just beginning to emerge. 

Insurance analysis in IoT 

For this next generation of insurance applications to be successful, analysis of the sensors and all other streaming data will be necessary – to understand existing conditions and to postulate future scenarios accurately. How will this work?

The stream of incoming data first should pass through an initial set of rules and algorithms devised for assessing the data, normalizing it as necessary and evaluating risk, continuously. The risk patterns are already defined to recognize when conditions are met in real-time. How? The pattern resides in the system as data is continuously streamed through – calculating the conditions to new events. When conditions are met, a reaction is triggered, sending the appropriate person or system the instructions to take the next action. For example, if vehicle monitoring indicates speeding based on location-based limits with frequent acceleration followed by sudden breakeage, an alert can be sent to the driver to reduce speed.

The high-volume and high-velocity granular data captured and transmitted from sensors and devices and attached to insured risks needs to be cleansed, organized and stored in a multi-parallel processing (MPP) data warehouse with effective history management. The stream of incoming data first should pass through an initial set of business rules and algorithms devised for continuous monitoring of risk. Any trigger of such rules should be able to send an appropriate feedback or control to the customer through state-of-the-art event stream processing system. The incremental data needs to be extracted from the warehouse and loaded into analytical input datasets for both modeling claim frequency and claim severity. The MPP environment supports high performance advanced predictive analytics and is used to discover any new significant rating parameter(s). The newly generated algorithm is then registered in a model management system and operationalized in the dynamic pricing engine.


The flow of sensor data for analysis in the insurance industry.

Unlike the current industry adoption of analytics in limited parts of the business, this new, IoT driven, big data revolution will require mainstream adoption of analytics across the entire insurance value chain. Every part of the business will be supported by sophisticated algorithms – assessing the current risks and calculating the differential value. Insurance organizations will be on the forefront of technology adoption to invest in scalable and robust high performance platforms.

Post a Comment

What keeps an insurance CEO awake at night?

I cannot speak from experience, but predominately an Insurance CEO has three primary objectives:

  1. Grow the business
  2. Reduce expenses
  3. Ensure compliance.

Let’s individually consider each of these objectives in more detail.

 Grow the Business

  • How does an insurance company grow from a $2bn to a $3bn organization? Essentially, insurance has a defined market size. Who needs a second home insurance policy or another life policy? Therefore to grow the business means increasing market share. To help achieve this objective, insurance companies are turning to analytics to improve pricing accuracy, create new products (telematics etc.) and enhance customer experience.

Reduce expenses

  • An insurance company has many different expenses. There are operating expenses such as employee salaries and infrastructure costs, underwriting expenses, and commission payouts. But by far the biggest expense within a property and casualty (P&C) insurance company is claims. Claims payouts and loss-adjustment expenses can account for up to 80 percent of an insurance company’s revenue. Adding analytics to the claims life cycle can deliver a measurable ROI with cost savings and increased profits; just a 1 percent improvement in the claims ratio for a $1 billion insurer is worth more than $7 million on the bottom line.

Ensure compliance

  • Insurance companies around the world are facing a host of new regulations. European insurers are consumed with implementing Solvency II, which will take effect in 2016. Many other countries are closely following events in Europe and seeking to implement the equivalent of Solvency II in their own regions. In the United States, the National Association of Insurance Commissioners (NAIC) has introduced the Solvency Modernization Initiative (SMI). To ensure compliance, insurance companies are implementing a comprehensive risk management frame work that includes data management, analytics and reporting.

In today’s highly competitive market, it is vital for insurance companies to minimize inefficiencies and reduce losses to protect profitability. By using analytics, insurance CEOs more easily achieve their objectives and sleep at night.

To learn more about how analytics can help, download the white papers “The Analytical P&C Insurer” and “The Analytical Life Insurer.”


Post a Comment