Insurance companies talk about new product development and speedto market, but true innovation is rare in the insurance sector.However, connected, communicating devices, often referred to asInternet of Things (IoT) could be one technology to revolutionizethe industry.

Sensors offer unprecedented access to granular data that can betransformed into assessing risk more accurately. For many insurerstheir initial exposure to IoT has been via telematics devices. Buttoday sensors are used in thousands of different devices. Sensorsare used in buildings and bridges to monitor for structural defectsand mitigate potential losses. Life insurance companies are usingthe data from wearable devices like FitBit and Nike+ FuelBand tobetter assess the health of the life insured. And, sensors arebeing implanted into animals to track and identify livestock,helping insurers rate and price agricultural insurance moreaccurately.

Related: 4 steps to building a strategic analytic culture inyour organization 

The first sensors appeared decades ago, but in the last fiveyears two major changes have shaken the sensor world and caused theIoT market to mature. From a technology perspective, the size andcost of the devices have decreased dramatically, and Wi-Fi andwireless communications make it more efficient to transmit thedata.

In an industry that's frequently slow to adopt cutting-edgetechnologies, IoT is starting to make waves. To successfullyleverage IoT, insurers need to invest heavily in both datamanagement and data analytics.

Data management

Big data has become a technology buzzword, and it is at theheart of IoT. First of all, let's consider the amount of data thatautomotive telematics devices are expected to generate. Atelematics device will produce a data record every second. Thisdata record will include information such as date, time, speed,longitude, latitude, acceleration or deceleration, cumulativemileage and fuel consumption. Depending on the frequency and lengthof the trips, these data records or data sets can represent up to 1GB of data per day, per vehicle!

To store this data, many insurance companies use distributedprocessing technologies such as the Hadoop file system. Hadoop isan open-source software framework for running applications on alarge cluster of commodity hardware. Since Hadoop runs on commodityhardware that scales out easily and quickly, organizations are nowable to store and archive a lot more data at a much lower cost.

To help insurance companies address thechallenges from large data volumes generated by IoT programs, it isessential for insurers to implement an enterprise data managementstrategy. This data management strategy should provide a unifiedenvironment of solutions, tools, methodologies and workflows formanaging telematics data as a core asset. It should also beflexible and scalable to reduce the time and effort required tofilter, aggregate and structure the exponential growth in data.

Data analysis

With all this new data that's available through IoT, how doinsurers determine which rating factors are predictive. Forexamples which data variables can forecast driving behavior,structural defaults or healthy living.  

Related: 3 ways the 'Internet of Things' will improvecustomer relations

The challenge is how to filter the noise from the signal. Addinga new variable increases the number of data points andrelationships exponentially. In a very simplistic model, if you aretesting for relationships among any five variables, there are 10two-way tests to run, shown in the equation (5×4)/2 = 10. If youdouble the number of variables to 10, you more than quadruple thenumber of relationships to test, shown by (10×9)/2 = 45. With IoTsensors adding dozens, if not hundreds, of new variables, thiscreates the potential to analyze millions of relationships.

That's big data analytics! The problem is that many of thoserelationships may be redundant or trivial, and hidden among themare the "real nuggets," or "signals."

The science of extracting insight from data is constantlyevolving. Tools are more readily available, and industries arebeginning to invest in the technology that supports big data. Byusing data exploration and analytics, insurers will be able to rankand weigh hundreds of new variables to develop highly accuratepricing models.

Figure: An example of a correlation matrix showing variablesfor auto insurance. Click image to expand.

Insurance companies cannot rely on traditional data miningtechnology to analyze all of this new data. Due to the sheer sizeof the data generated by sensors and IoT, insurers must consider adistributed, in-memory environment to display the results of dataexploration and analysis in a way that is meaningful but notoverwhelming.

Exploiting the Internet of Things

The IoT has the power to transform many aspects of the insuranceindustry and deliver significant competitive advantages to earlyadopters. But with this great potential also comes complexity.Advanced high-performance analytics and big data tools can assistcompanies in overcoming the complexities, enabling them to reachthe full potential of IoT as it grows from a trend to a must-havefor all insurers.

Related: Beyond big data: Inside the technological evolutionof property & casualty insurance

Stuart Rose is global insurance marketing manager at SAS. Hebegan his career as an actuary and has over 20 years experience inthe insurance industry.