The last word on outsourcing

Last week's column about the commoditization of IT services elicited spirited responses from IT professionals once again, especially those working in desktop support. (To those of you thinking, "Gosh, I really wish this Dickerson character would get off his desktop support kick," note that this time I'm using the topic as a springboard to loftier ideas.) One reason desktop support sparks interest is that it's a universal IT problem. At the same time, conventional wisdom among IT professionals dictates that it's a hands-on, in-house job.

Free IT resource Hear how top CIOs turn change into a competitive advantage. Sponsored by HP Free IT resource Try Sun servers, workstations and storage products free for 60-days. Sponsored by Sun Microsystems

That's where I profoundly disagree -- and where the more general lesson of IT comes in. That lesson is simple: If you think a well-defined function can't realistically be automated and then outsourced, chances are that you are wrong. Nothing in IT is immune from the march of automation, which after all is the heart and soul of IT.

In making my argument that desktop support should be outsourced, I haven't been clear about the specific model I'm advocating. There are two distinct outsourcing models, one of which I see as clearly superior to the other, both financially and technically. The first is a people-focused direct staffing model where the outsourcer simply hires and places staff on-site to do the same job IT staff on the company payroll might have done and using the same old methods: reacting to problems as they arise and managing a queue of help tickets, usually within strictly enforced banker's hours. I have never considered this model because it seems more expensive and just as labor-intensive, providing no obvious advantage to using your own staff.

The other model is the outsourced remote support model, which is what I use at InfoWorld. Leveraging nearly unlimited bandwidth, this technology-focused model allows SMBs like InfoWorld to enjoy technological economies of scale that are simply impossible for most SMBs to replicate at a reasonable cost. The basic desktop support issues are simple: virus-free desktops, proactive patching of Windows systems, guarding against filled-up disks, backups/restores, and software distribution. Reliable -- but complex and expensive -- software solutions exist for all of these functions, of course.

But can your company afford to buy the software, spend the IT staff time to test it, and integrate it into all offices from a central location? Can you really manage all desktops in all locations through a remote-control mechanism? Are all your desktops being patched automatically with well-tested patches and backed up daily, regardless of location? Is there an institutional mechanism to measure end-user satisfaction across the board? Can you provide a uniform level of 24x7x365 support? I can.

E-mail responses from some of my fellow IT pros suggest that my advocacy of this sort of outsourcing is traitorous and anti-IT. To them I say: Get over it. The economies of scale have settled the question and have established outsourcing as part of the overall solutions mix. Nothing is sacred in a discipline built on automation. That doesn't mean, of course, that it's smart to offload any commoditized function to the nearest provider. I won't consider outsourcing unless my outsourcing partner is leveraging technology that provides me with a reliable, less expensive, and -- in most cases -- better solution. Based on those simple criteria, the decision to outsource desktop support wasn't difficult for me. I'm not sure what function is next, but I'm certain this story is not over.