One of the more interesting characters at EMC is the Dean of Big Data, Bill Schmarzo. When Dr. Jim Short and I kicked off the Architecting for Value research in 2015, Bill was one of the first people we spoke to. Bill has his finger on the pulse of data valuation due to continual contact with customers, many of whom are trying to extract value out of their data. This summer Bill wrote an article sharing his own personal views on calculating the economic value of data. He has also identified a Rubik’s cube of three sources of intellectual capital that contribute to data value:
- Data
- Analytic models
- Decisions (or business use cases)
Bill has neatly depicted the relationship between these three as part of his Big Data Intellectual Capital Rubik Cube blog post. I’ve pasted this diagram below:
Bill is finding an immediate need for corporations to formally track their data assets and analytic models. When looking closely at the diagram above, the bottom right-hand side depicts a Data Lake framework. Building such a framework allows for a corporation to maintain both a data catalog as well as an analytic catalog. Bill’s approach suggests that the assets (the data and the models) located within the Data Lake be directly applied against business decisions with known (or estimated) financial or revenue impacts. This approach is consistent with the research on Digital Business Platforms (DBP) being conducted by Wikibon, in which data is viewed as a capital asset that is used to drive corporate revenue alongside more traditional assets.
Pulling together these three pieces of IP and running it on a DBP framework (such as a Data Lake) allows an individual corporation to formally use data as an asset to drive business results. This practical approach can now be compared against some of the advanced use cases Dr. Short and I have been seeing in the industry, including:
- Insurance and Data Value. How can the process of cyber-insurance become more automated for both the insurer and the insured?
- Valuation and the Sale of Data Assets. Can a DBP be used to identify and prepare assets for sale?
- Data Monetization. Can a DBP identify assets that can be “rented” for profit?
- Data Value and Bankruptcy. Does asset management of data and data models better enable bankruptcy use cases?
- Data Value During Acquisition: Does a company become a better acquisition target if their house is in order from a data and model point of view?
Each one of these advanced use cases represents a “stretch goal” that companies can address once they begin to master the recommendations being presented to the industry by Bill. I hope to dive into the implementation of these use cases in future posts. Until then, I strongly recommend viewing Bill’s interview with Peter Burriss of Wikibon Weekly (embedded below):
Steve
EMC Fellow
