Deflation and Cloud Commoditization at the DoD: Best Practices
Published: August 26, 2015
Prices for cloud services are dropping and with its new best practices guidance the Department of Defense is positioning to take advantage of it.
Am I mistaken or is there an extremely important word missing from the Department of Defense's newly released Best Practices Guide for Department of Defense Cloud Mission Owners? The word I’m thinking of is “competition.” According to what industry and Congress have been hearing from Defense officials for years moving to the cloud offers cost savings that the department cannot get via other acquisition avenues. Of course there are other benefits, but reducing costs has long been cited as one of the biggest. Reading through the new best practices guide, however, I’ve been struck by the fact that there is no mention of how to conduct a cloud contract competition under the new acquisition paradigm. There is plenty of advice on determining bandwidth needs and on selecting a commercial provider from the list of approved Cloud Service Offerings, but there is no advice on how DoD Mission Owners are supposed to select approved cloud providers and compete a contract among them.
Perhaps this process is locked away in the file cabinets of the department’s contracting offices. If it is, I apologize for my comments. However, given that agency acquisitions are supposed to be fair and open maybe I can be forgiven for assuming that guidance concerning the competition of cloud contracts within the new paradigm would be open and publicly available.
Since this guidance hasn’t yet been made available, let’s carry on under the assumption that there is no new guidance about how to compete contracts among approved cloud providers. Cobbling together the best practices guidance and pieces of evidence from other bits of the DoD’s cloud acquisition policies, the picture that emerges is of an acquisition environment constructed to be like a retail store. The Defense Information Systems Agency (DISA) compiles the approved services into catalog. The catalog’s offerings are organized according to data impact (i.e., security) level and service delivery type. Then, Mission Owners work with DISA to match their requirements with the commercial provider that best suits their needs.
This paradigm raises a host of questions. Is the commercial provider selected the only one available? Is the contract awarded for the service based on a previously submitted table of rates? Are Mission Owners required to find three commercial providers to compete rates among? Will a solicitation be released on Federal Business Opportunities to secure proposal responses permitting a formal evaluation? There are other questions as well, but you get the picture.
The point in asking these questions is understanding exactly how the DoD can secure the best price for its mission needs when there doesn’t appear to be a competition.
The answer appears to lie in the commoditization of cloud services, which I’ve been writing about for some time now. This trend has been developing in the commercial environment and it shows no signs of abating. The result of it is that prices for cloud services are dropping, meaning profit margins are also falling. Vendors will need to make up for the reduction in margin by generating volume. Basically, the market is in a deflationary environment being driven by technology and the DoD is positioning itself to take advantage of it. Cloud capabilities may be described as “services,” but the DoD’s approach seems to treat them like commodities and commodities are subject to a variety of pressures that drive down prices.
Back in 2010, Vivek Kundra estimated that the federal government could shave $20 billion off of its IT spending by moving systems to the cloud. Industry appears to have taken that as meaning $20 billion would be shifted to spending on the cloud. But was this actually what Kundra meant? It’s possible he meant the federal government could spend as much as $20 billion less on IT goods and services by moving to the cloud.
Twenty billion dollars less. That would shrink the federal IT market to $66 billion per year, using OMB's baseline estimate of annual IT spending. What would a market like that look like?