Mine Social Media Data to Identify, Manage Risk

by   |   August 25, 2014 5:30 am   |   0 Comments

Mark and David Mine Social Media Data to Identify, Manage Risk

Mark Pluta, left, and David Ahrens, STARS

Engagement with customers through social media has become an essential element of most integrated marketing campaigns. Marketing organizations are keenly focused on visualizations of social data and reaching influencers who impact the preferences of others in their networks. As part of brand management, marketers must hear and be heard on multiple social media platforms.

Risk managers should join marketers in keeping their ears to the ground (or perhaps we should say ears to the cloud) to hear what’s being said about their companies on Facebook, Twitter, LinkedIn, and other social networking sites. Monitoring and utilizing social media should be key components of a company’s overall risk-management strategy. The following are specific examples of how risk managers can leverage social media data to gain insight that will help reduce their total cost of risk.

Related Stories

Assess and Manage Data Risks with Total Information Risk Management.
Read the story »

Establish Electronic Discovery Policies as a Risk Management Strategy.
Read the story »

How Data Profiling Improves Management of Storage Media, Risks, Costs.
Read the story »

Brand Owners Use Location Analytics Lens to View Social Media Streams.
Read the story »

Dynamic Risk Management. The speed with which an incident is identified and addressed can have a significant impact on loss severity. Monitoring social media gives risk managers information in real time so that they can respond and potentially limit the effects and scope of the incident. In 2009, a group of researchers from the University of Iowa collected and stored public tweets related to the H1N1 influenza. Among their findings was that Twitter traffic could be used “to estimate disease activity in real time, i.e., 1-2 weeks faster than current practice allows.”  In a 2012 paper, two researchers from the University of Tokyo examined how Twitter, as a real-time means of communication, could be used for earthquake detection. The authors regarded each Twitter user as a sensor and each tweet as sensory information.  Given the importance of sensors to the Internet of Things, a key enabler for dynamic risk management, the authors’ vision of Twitter users as “social sensors” is insightful.

Identifying Location-Based Risk. When companies identify where incidents are most likely to occur, they can focus resources and allocate costs appropriately. Monitoring social media can inform companies of safety issues or the potential for incidents in certain global locations. Law enforcement agencies are using social media for crime mapping and identifying geographic locations that are at highest risk. Predictive policing is now being used by a number of police departments, including the Los Angeles Police Department. According to a 2013 social media survey by the International Association of Chiefs of Police, 95.9 percent of agencies surveyed used social media in some capacity, including 66.1 percent that used it for intelligence and 45.7 percent that used it for listening/monitoring.

Identifying and Mitigating Reputational Risk. What’s being said about a company on social media, whether by customers, employees, or competitors, can erode the company’s reputation and taint its brand. A notable example is Domino’s Pizza, which saw its reputation severely damaged in 2009 after an unsavory online video posted by two Domino’s employees went viral. Many companies learned from this high-profile incident and instituted social media strategies. Monitoring social media can alert companies to reputational hazards so that they can respond quickly, before the situation spirals out of control. That response should include social media channels in addition to other methods so that the online community recognizes that the company is taking action to remediate the problem.

Determining Risk Exposure. Companies need to know whether one person or hundreds are blogging or tweeting about a product defect. It’s also important to know whether a given blogger is an influencer (influencers are people who others follow or whose opinion influences others). Social media can deliver an early warning for a product design or performance problem (many remember the antenna issues with the Apple iPhone 4, later referred to as Antennagate), and give companies the opportunity to identify safety and performance issues early on. For example, a case study conducted by researchers at Virginia Tech’s Pamplin College of Business showed that social media could help automobile manufacturers uncover and categorize vehicle defects before they appear in National Highway Traffic Safety Administration (NHTSA) communications.

Identifying and Responding to Crises or Disasters. Monitoring social media is an important tool to maximize situational awareness during and after a crisis or disaster. Particularly when other methods of communication are compromised, social media is essential for companies to be alerted when an incident occurs and stay informed of events and their locations in real time. A paper published by the Organization for Economic Cooperation and Development in 2013 describes how social media improved situational awareness for responders after the 2010 earthquake in Haiti. When telephone landlines were unavailable, Haitians turned to social media on their mobile devices. Social media data was aggregated by responders for interactive mapping. Of the organizations participating in the Continuity Insights “Crisis Communications: Social Media & Notification Systems” survey conducted in 2014, nearly 62 percent planned to use social media to gather information during an event or crisis, but only 37 percent used geospatial mapping to visually represent the location of employees, assets, and incidents.

Claims Fraud Mitigation. If an employee posts a selfie on a ski slope shortly after submitting a claim for a leg injury, it could be a sign of fraudulent claims activity. Earlier this year, more than 100 people were charged in a major Social Security disability fraud case, including former New York City police officers and firefighters. A January 8, 2014, Wall Street Journal article about the case features a photo of an ex-police sergeant who claimed that he couldn’t leave his house enjoying a fishing trip. Social media provides important clues for investigators as they seek to identify questionable claims. For example, geotagging can provide essential information about the location and date associated with photos posted online. Social media can also be used to detect identity fraud and organized fraud rings among individuals in a claimant’s social graph.

For risk managers, social media introduces a host of new risks that need to be anticipated and managed. At the same time, social media promises to be an important tool for managing risk. The key is incorporating social media strategy in a company’s risk management initiatives. Risk management software that features social media monitoring capabilities will facilitate this integration. In the 2014 Continuity Insights survey, less than 3 percent of the companies surveyed gave control of social media to their business continuity organizations, compared to 41 percent that gave control to corporate communications organizations. To have full visibility of factors impacting their operations, risk managers need to be active participants in their companies’ social media activities, playing both offense and defense.

Mark Pluta is CTO and David Ahrens is CMO at STARS (a division of Marsh).


Subscribe to Data Informed
for the latest information and news on big data and analytics for the enterprise.


bac1ef92 d2e3 48e2 b6ee 9794d39d55e0 Mine Social Media Data to Identify, Manage Risk



Tags: , , ,

Post a Comment

Your email is never published nor shared. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>