What Were the Results and Their Meaning or Context?
How Are These Results Meaningful for Organizations and for Future Research?
The data are very revealing. On just the level of CI activity per industry sector, there is some evidence of more activity when higher levels of Big Data are present with double-digit CI activity scores. Essentially, there are more, and more professional, CI operations in data-heavy industries. But these metrics are also less than consistent with the highest levels of activity closer to the middle Big Data. There are likely explanations as we’ll see, but these initial results are also a reason to look more carefully at the data.
Based on the additional adjustment, calculating the per firm CI metric to go with the per firm Big Data metric makes sense. The number of firms in the extremely broad MGI industry classifications don’t necessarily match up well with the number in the SIC categories for IC. Indeed, if one closely examines the data, some industries with relatively high numbers of CI activity also have an abnormally large number of firms represented (retail, wholesale). Similarly, some of those with seemingly muted CI numbers have a relatively low number of resident firms
(securities, utilities, natural resources). Consequently, the per firm CI score not only provides more information but is probably the truer representation.
And it does provide some very interesting results. Just as the stored data per firm column shows a clear progression from top to bottom, so does the CI activity per firm column. The level of CI activity for industries with a commitment to Big Data varies within a fairly small range but is of a magnitude difference from the level for industries with low commitment. Scores in the single digits and low double digits (showing aggressive CI operations) go along with a high degree of data storage per firm. Scores in the triple digits (muted CI operations) go along with a low degree of data storage per firm. The pattern is clear and convincing. High levels of CI activity appear to go along with high levels of investment in big data capabilities. Big Data has the potential to feed competitors’ CI operations, and those pursuing Big Data strategies should probably be sure to invest in data protection, counterintelligence, and their own CI operations.
But there is even further depth to the data given what we know about KM/ IC and CI in some of these industries. Those with the absolute lowest CI ratios— insurance, utilities, and securities—are all regulated industries generating a lot of operational and transactional data. Not all of that data is particularly interesting (movements of money, investments, power). Indeed, what we know from KM/IC is that these industries do not rate very highly for intangible assets, period (Erickson and Rothberg 2012). What is valuable is the rare tacit insight, the “eureka” moments that result in new directions in such industries (new portfolio ideas, new lending or investment strategies, whatever is new in today’s old-line utilities). In such environments, the ability of CI operations to spot new ideas from competitors, based on changes in Big Data patterns or discovering the insight itself, can result in rapid copying. Consequently, CI can be both effective and profitable by cutting down periods of new product or new process exclusivity.
Similarly, the next lowest CI ratios are in manufacturing industries, both process (pharmaceuticals, plastics, chemicals) and discrete (machinery, transportation). Again, there is value in CI, not necessarily from the supply chain or operations data, but from what it tells observers is going on in terms of R&D and new products, process optimization, or other manufacturing improvements. Further, just as in the previous group, actual activities tend to be hidden from view, so CI operations of a certain maturity are needed to peek behind the veil.
At the other end of the spectrum, a very different pattern emerges. Those with very high scores on the metric (indicating low levels of CI activity) are invariably services. Services can employ Big Data as they have operations, transactions, and communications just as with other industries. Services, however, have unique characteristics that make them hard to manage, including intangibility, perishability, producer variability, and customer involvement. Consequently, optimizing and standardizing processes can be difficult, leaving a gap in terms of what Big Data is able to accomplish. Moreover, the services seen here are often right out in the open, making advanced CI operations something amounting to overkill. Much of what can be learned could come simply from walking through a public facility and observing the layout or operation. In such situations, fewer worries exist about CI, so there is less need for protection (or, as pointed out, important matters are so transparent as to be almost impossible to protect). There also appears to be less need for counterintelligence or heavy investment in a CI operation.