Let’s talk about ITIL.
Regular readers of this column might be under the impression that I’m hostile to the very notion of ITIL. But I’m not. It’s a rich source and organized catalog of ideas and techniques for running IT processes effectively.
Not that I’m an unquestioning admirer. Based on what I’ve learned about the subject, ITIL has at least four major weaknesses:
- Process-centricity: ITIL is a process framework. Processes constitute only about a third of the factors we’ve mapped that IT needs to master, and not the most important third, either.
- Processes vs practices: Regular visitors will recognize that processes and practices are different animals. ITIL fails to distinguish between the two, resulting in excessive emphasis on the steps to be followed, and too little on the excellence of the practitioners.
- Personal technologies: ITIL doesn’t appear to recognize that how IT manages personal technologies … personal computers, telephones, smart phones and so on … is a separate and distinct discipline that has a disproportionately heavy impact on IT’s overall effectiveness in the enterprise.
- Embrace of the “Standard Model”: The “Standard Model” of information technology organizations is that they’re supposed to (1) be run as a business, that (2) delivers working software, to (3) its internal customers. ITIL doesn’t officially require the standard model, yet the Standard Model lurks inside it in the form of definitions and both hidden and overt assumptions. For most IT organizations, everything about the standard model is a problem.
I’ve written these criticisms as definitive statements. They are somewhat less than that. ITIL’s so-called “best” practices are really lists of alternative practices and/or processes — each workable, none “best,” and without much to help you decide which fits your specific situation.
For example: I recently criticized ITIL for endorsing IT charge-backs as industry best practice (“Pentagonalitil,” 8/3/2009). I based my statement on materials embedded in a number of ITIL consulting and support software vendor sites.
Not so, I was informed by regular correspondent Dave Ivey, who informs me that, “Section 5 of the ITIL Service Strategy book addresses Service Economics. Section 5.1.4.1 states that the real issue is not whether to treat service as a cost or profit center, but rather how an organization chooses to replenish the funding of its service operation.
“Section 5.1.4.2 is titled ‘Chargeback: to charge or not to charge.’ Considerations for both are given, and various models are explained. The section ends with: ‘No matter which methodology is used, or none at all, it is more important to make certain that the overriding substantiation comes from providing value to the business.'”
Mr. Ivey also informs me that ITIL’s account is descriptive rather than prescriptive: It describes various charge-back models, but doesn’t provide guidance as to how to match circumstances to model. (To be fair, advising that, “Charge-backs are best-suited to companies with centralized IT, a high level of politics, limited leadership, and ineffective governance,” might be considered insufficiently diplomatic.)
ITIL’s greatest strength is that it provides a comprehensive and well-organized list of practices that might not be best but are at least workable. It’s most annoying characteristic is that its purveyors require commitment before providing enough information to warrant it. ITIL Service Strategy alone costs a spendy $139.40. At 276 pages that’s fifty cents a page. (For comparison, the KJR Manifesto‘s per-page cost is a comparatively svelte nine pennies.)
If I were a suspicious sort, I’d think it’s a deliberate ploy to keep the riff-raff out.
So: High value, high cost, and significant shortcomings. If it were software, someone would be thinking open source. Which brings me to a modest proposal: Why not?
Why not create an open-source alternative to ITIL?
Consider: Open source isn’t limited to software. It’s a way to organize like-minded communities that create and refine pieces of intellectual property.
It might not be possible or practical. On the other hand, a well-constructed framework that:
- Takes into account not merely process but all of the factors needed for IT to be successful …
- Provides contextual guidance besides, to make clear where different solutions work and where they don’t …
- Rejects the “Standard Model” as an underlying premise (it could certainly accept it as a situationally appropriate alternative) …
… could be immensely valuable.
“All” that would be needed is a community willing to invest time and effort to make it a reality.
Any volunteers?