Gartner Blog Network

Why Government IT Consolidation May Be Doomed

by Andrea Di Maio  |  March 9, 2012  |  1 Comment

Yesterday I spent the whole day with government clients in a European country where there is an aggressive plan to consolidate infrastructure across several departments. This is one of the many data center consolidation programs that are going on around the world in all tiers of government. Some of them are quite aggressive and are driven by a mandate, like in Canada, while others are more aspirations or part of a normal (and indeed healthy) rationalization process.

Now, who would be against the idea of consolidation? The number of data centers that many governments run and their low level of utilization almost cry for consolidation: benefits are clear, ranging from greater security to lower costs, from lower energy consumption to easier maintenance, and more.

We all know that there are a number of non-technology challenges, such as fear of losing headcount, budget, authority, but all this can be managed if there is sufficiently strong executive leadership.

However, there is a different, subtler risk to consolidation, and some of the other initiatives that governments themselves have set in motion amplify such risk.

Put yourself in an agency CIO’s shoes. You are told that data centers will be consolidated across government, and yours could be one of those that are going to be retired. As you need to support your agency business, you start exploring what options you have to minimize your exposure to a possible delay or failure of the consolidation program, or just to the emergence of new business needs that require changes to existing applications or entirely new ones. You need to reduce your cost base but also to be in control of your own destiny, and you know that you data center has an expiration date: therefore you would probably start looking at various sourcing options, one of which is the consolidated data center, but that would include also hosting, outsourcing and cloud sourcing.

And here comes the irony. Those same government organizations that are carrying on data center consolidation are also leading a cloud computing effort, that often leads to providing procurement vehicles that are meant help whoever is willing to use cloud., the UK CloudStore, the NZ panel contract for IaaS, and more to come give you as the CIO a lot of choice about how to meet your agency’s business needs. As the maturity of these cloud stores improves, your business users could buy straight from the cloud store, bypassing (within reason of course) their IT organization – i.e. you.

One might argue that providing more choice to individual departments and agencies is a good thing (and I personally agree that it probably is, assuming they can use it judiciously). However what this does is to undermine the business case that was behind the data center consolidation initiative.

Let’s take a hypothetical example. Let’s say that there are about one million government staff that are the target users of an IT consolidation initiative. The savings are predicated on the assumption that one million people’s worth of workloads will be hosted in the new consolidated data centers.Since this migration cannot happen instantly but takes years (at least two if you are an optimist, most likely much longer), in the meantime a number of agencies, accounting for – say – 200,000 staff may decide – over time – to move into a variety of cloud offerings. This means that the target for consolidation is now 800,000. And of course, if the migration of those 200,000 is successful, others are likely to follow, further decreasing the target for the consolidated data centers. The trouble is that the original business case that justified the consolidation was based on one million users, and it may be hard to demonstrate that the case still holds with 20 or 30% less. Of course number of staff is only one possible metric, and one could consider number of applications or workloads or transactions and so forth.

And make no mistake: even when consolidation is mandated by taking over IT budget and staff, nothing prevents the business from looking into business process as a service offerings, hence circumventing the mandate while the consolidation is still ongoing and equally hurting the business case.

The bottom line is that consolidation still makes a lot of sense, but requires a far more realistic business case.

Category: cloud  it-management  

Tags: consolidation  shared-services  

Andrea Di Maio
Managing VP
15 years at Gartner
28 years IT industry

Andrea Di Maio is a managing vice president for public sector in Gartner Research, covering government and education. His personal research focus is on digital government strategies strategies, Web 2.0, open government, cloud computing, the business value of IT, smart cities, and the impact of technology on the future of government Read Full Bio

Thoughts on Why Government IT Consolidation May Be Doomed

  1. MikePearsonNZ says:

    I think Government IT is doomed, where it involves commodity IT like severs, storage and pcs; but Government IT trusted advisory services will take off. There will be a role for trusted advisors to help the gov business how to work within gov requirements. The people holding those internal positions may think they are better placed to provide those services, but it’s not a foregone conclusion — IBM, etc strategically positioned themselves over 10 years ago now, could they offer the service as well?

Comments are closed

Comments or opinions expressed on this blog are those of the individual contributors only, and do not necessarily represent the views of Gartner, Inc. or its management. Readers may copy and redistribute blog postings on other blogs, or otherwise for private, non-commercial or journalistic purposes, with attribution to Gartner. This content may not be used for any other purposes in any other formats or media. The content on this blog is provided on an "as-is" basis. Gartner shall not be liable for any damages whatsoever arising out of the content or use of this blog.