Log in Join now News & media

Risk management needs "smart data" not big data

Published on Fri, 24/05/2019 - 16:59
Luke Anear, CEO, SafetyCulture
Luke Anear, CEO, SafetyCulture

While predictive risk models may dominate headlines, Luke Anear, CEO of SafetyCulture argues that collecting real-time data at the point of risk is the key to making an accurate assessment of risk.

Data and the potentially powerful insights that can be extracted from it is today widely recognised as being essential for all those involved in the management of risk. For insurers, it helps them manage their loss exposure across portfolios, and for businesses, not only does it help manage risk, it can also result in preferential policy terms.

However, for most businesses, knowing what data to focus on and how to extract value from it is a challenge. If data is good, large volumes of "big data" is often assumed to be better.

However, a greater quantity of data does not necessarily lead to better decision making. In the rush to find more data, it can be easy to overlook more fundamental criteria that can help with better risk management and selection. In the words of Albert Einstein: "Everything should be as simple as it can be but not simpler".

It is in this respect that data captured by those working on the frontline should not be underestimated. The ability to collect accurate, real-time data by those operating at the point of risk has presented the next step in the evolution of risk management; workers collect and submit accurate data in real-time, the output of which is centrally analysed and actioned by management.

The most effective ways of measuring risk are often the simplest

The explosion in the volumes of data available to risk management professionals has seen the emergence of hundreds of new companies offering to create unique analytical insights for insurance companies and their clients. Many present artificial intelligence, the "internet of things" and blockchain as the core of their solution. While these technological developments are important, none can replace approaches which focus on data accurately captured at ground level

Access to high quality, targeted data is important. Often the most effective measures for identifying risk are the simplest and in the search for the smartest application of artificial intelligence, it's important not to overlook the simpler, but often more telling signs that point to the quality of risk management (both good and bad).

Take risk-based checklists as an example. In the last thirty years, the use of pre-flight checklists has become an essential part of the process used by every aircraft pilot. Checklists have become increasingly common in many industries, usually in response to regulatory requirements. Until recently these were paper-based, carried out infrequently and generally only used by those with a specific health and safety responsibility.

The idea of using tablets or mobile phones to create digital checklists has been around for a while, but until recently most applications have been specific to particular industries or companies, with little or no sharing of data.

Audits as a lead indicator?

This is changing. The introduction of mobile inspection solutions are now starting to become available and used across all types of industries to assist with audits. The data shows that the extent to which companies effectively embed inspections into their work practices and create a culture of risk management has a high correlation with lower accidents and losses.

A handful of insurers and brokers have recognised that the use of audits is a leading indicator of a lower risk profile between companies in comparable industries. Some insurance companies are now using mobile risk inspections to monitor potential risks throughout their clients' organisations in a fraction of the time it would traditionally take.

This process digitises data collection to allow for greater collaboration between the insurer and the insured, and ultimately leads to new ways to understand, manage and price losses whether physical or intangible. The sharing of quality, validated data allows for greater alignment and shared interests for both parties. We will move from a world of "big data" to "smart data".