Last week I discussed Data Valuation with the IT@Cork community in Ireland. During this talk I introduced data valuation in the context of data insurance. In order to insure data, someone has to assess the data's economic value. By exploring this use case I believe that insight can be gained into the people and processes involved in data valuation. I also believe that understanding the people and processes can illuminate the role (if any) of IT infrastructure in data valuation.
Several months ago a local paper wrote an article on the rise of data insurance. As part of that article the author cited three examples of negative valuation:
In the three cases above, the "negative value" of customer records can be calculated as follows:
- South Shore Hospital: slightly less than one dollar per record
- TJX: slightly less than four dollars per record
- Sony: 1.71 dollars per record
Would it have been possible, before these breaches, for a corporate data analyst to predict the negative economic value their company would experience in the case of a breach? How accurate would this prediction have been?
There are many such examples of negative data valuation. Accordingly there are a large number of examples of positive data valuation. In a discussion with my EMC colleague Ed Walsh he pointed me to recent news about a personal genetics company 23andMe. The article states the following about 23andMe's data sets (which are generated via DNA test kits).
Since it was founded in 2006, 23andMe has collected data from 800,000 customers and it sells its tests for $99 each.
At $99 per kit this data has a baseline valuation of just under $80 million dollars. Over time, however, the corporate value of this data has increased. According to the article, at least one company is willing to pay top dollar for access to the DNA data:
According to sources close to the deal, 23andMe is receiving an upfront payment from Genentech of $10 million, with further milestones of as much as $50 million. The deal is the first of ten 23andMe says it has signed with large pharmaceutical and biotech companies.…this single deal with one large drug company could generate almost as much revenue as doubling 23andMe’s customer base
For both the positive and negative aspects of data value, the slide below lists a set of questions that are worth asking about the process of data valuation:
In future posts I will explore these questions.