Pinpoint positioning through geospatial and big data

There's been a lot of talk about ‘big data' since McKinsey published its Big data: The next frontier for innovation, competition and productivity report last May.

McKinsey wrote: "The amount of data in our world has been exploding, and analysing large data sets, so-called big data, will become a key basis of competition."

The company explained that the insurance sector could benefit from big data, as long as it was able to overcome barriers to its use. With this in mind, Ordnance Survey recently compiled research, in conjunction with IDC, to find out what insurers need to do to unlock this powerful mine of information.

Insurers churn through escalating volumes of transactional or event data every second, creating trillions of bytes of information about their customers, brokers, clients and suppliers.

Real-time event and transactional data may include implicit location data: addresses, place names or regions. Some mobile and sensor event data may include explicit location data: coordinates, elevations, building footprints or shapes.

The key to bringing vast data sets together to create a single customer view is the ability to transform between these two data types.

For example, underwriters need the precise location (explicit) of the property they are insuring, right down to co-ordinates. Marketing managers will use regional or district location (implicit) data to identify areas with a common profile to target, and risk managers need to understand the accumulation within a given area.

Customers using smartphone claims apps already have the ability to pinpoint pictures and their account to the exact time and co-ordinates, or explicit data, of an accident. Combining and having the ability to switch between the data sets will provide competitive edge.

Taking this location, or geospatial, approach is different from traditional methods, which focus on documents first to analyse data and isolate keywords to understand meaning or sentiment. Using a geospatial approach will enable the linking of disparate legacy systems by matching the key theme used across these systems: location.

But it's not enough to just match it. Analytics is at the heart of the big data revolution. By applying location to big data analytics, insurers can develop the capability to leverage more insight from their data and understand risk and customer behaviour in greater detail.

As technology moves towards real-time big data analytics, most analytics will have to be explicitly location-specific. Two key areas driving this need for real-time analysis are underwriting and fraud.

Underwriters are desperate for a real-time analysis of their cumulative exposure, and fraud managers are keen to spot fraudulent claims in real time using location-specific analytics.

But insurers will need the right type of talent to make the most of data analytics, and McKinsey forecasts a shortage of analytics graduates within five years.

The opportunities presented by big data are clear: improved productivity, growth and innovation providing a competitive edge. It will be interesting to see which insurers take the lead in ensuring that the true power of data is understood.