Those who have been watching have witnessed the growth of the application of predictive analytics in a variety of business operations.
The use of predictive coding was discussed in a recent WSJ blog post, CIOs Aren’t Leveraging Predictive Software Enough, Says Gartner, but it had nothing to do with ediscovery (believe it or not, there is a world outside of TAR). Marketing analytics was the focus of the piece, but the objectives are conceptually very similar: use sophisticated analytical technologies to learn subtle patterns from cases of known data to make predictions about “unknown” data.
Those who have been watching have witnessed the growth of the application of predictive analytics in a variety of business operations. And we’ve seen that because of the information load contained within unstructured organization information, text analytics and mining has become an important R&D area. The technologies underlying statistics-based predictive coding solutions in the ediscovery industry are but one branch of a growing tree of predictive analytics applications. There are many others. The inevitability of analytics in ediscovery is simply one example of a much larger reality: we are in the midst of a big data, big analytics revolution.
In Competing on Analytics: The New Science of Winning, Thomas Davenport describes the process of developing a strategic view for integrating analytics into the DNA of enterprises to provide them with the competitive advantages of insight, cost reductions and revenue growth. In part, it calls for an objective review of the areas where analytics can best be applied to secure those advantages, and how to develop solutions within a consistent integrated analytics environment.
Given ediscovery costs and the advent of predictive coding, the application of emerging technology-enabled ediscovery solutions needs to be part of that strategic discussion. Certainly, CIOs, working with the legal department, should be aware of predictive coding as an important part of the solution to legal spend, and should be part of the decision-making process as to whether and how to incorporate predictive coding internally.
But going beyond that, stakeholders need to begin thinking about if and how all of the analytical “IP” that is developed during the legal review process can be used and re-used by and for the legal department as well as other enterprise communities. And companies need to think about taking ownership of that IP.
Given the evolution of cloud services and providers, and the emergence of embedded analytics solutions, it’s a propitious time to contemplate whether onboarding predictive coding makes sense for the enterprise.
