Other parts of this series:
“By implementing the right structure, safeguards and transparency, insurers should be well positioned to succeed in their Finance and Risk big data journey.”
In this blog series exploring big data and the insurance industry, we’ve looked at the trends big data is driving, as well as a series of steps insurers can take to actually use big data to keep pace with those trends. The prior post looked at the “lessons” insurers need to learn—because working with big data isn’t the same as business as usual.
The first lesson, as noted in my previous post, focused on data governance. So what’s next?
Delivering safety: Data privacy and IT protection
Data privacy/data protection[i] is defined as the process of protecting all forms of data (personal or corporate) against internal (diversion, misappropriation…) and external (hacking, destruction…) threats. The efficiency of this process depends on the balance between individual privacy rights and the use of data for business purpose.
Source: Accenture, August 2016
Data privacy in a big data context should be viewed from three perspectives:
- Legal: A common denominator derived from European Union jurisdiction and proven principles to define data protection requirements (e.g. Directive 95/46/EC, General Data Protection Regulation (GDPR), Regulation (EU) 2016/679).
- Data governance: Analytics-focused data governance that translates data protection requirements into technical solutions.
- Technology (platform architecture, tool configuration and IT security): Technical solutions to facilitate data protection and regulatory-driven data compliance requirements.
Know your approach: Metadata management
Metadata (data about data) provides the additional information that helps humans and software make sense of data and derive consistent analysis and actions.
Metadata characteristics change substantially with big data, especially regarding its unstructured nature and speed of evolvement:
Source: Accenture, August 2016
To help win this challenge, insurers can take the following measures to make full use of data’s potential:
- Set up an “inference engine” that automatically extracts context and sentiment from data (e.g., to derive risks from social media data).
- Use an integrated, open metadata model (e.g., the common warehouse metamodel used across the data pipeline to exchange semantics across all applications involved in the closing process).
- Establish a communication strategy to distribute metadata to all applications, IT and business users to ease understanding of metrics (e.g., when creating financial statements or regulatory reports).
By implementing the right structure, safeguards and transparency, insurers should be well positioned to succeed in their Finance and Risk big data journey.
Continue learning about the big data transformation at: A New Approach to Data Management in the Digital Era.
[i] The usage of terms “data privacy” and “data protection” varies from country to country, depends on the language spoken in that particular country, regulatory regimes and media’s use of the terms.
 The usage of terms “data privacy” and “data protection” varies from country to country, depends on the language spoken in that particular country, regulatory regimes and media’s use of the terms.