I’ve written about metrics in terms of their general ability to help illustrate IT’s value to the business. That’s especially important for initiatives in which business people have a tough time seeing the connection to broader business goals and thus may resist making investments.
I’ve written about metrics in terms of their general ability to help illustrate IT’s value to the business. That’s especially important for initiatives in which business people have a tough time seeing the connection to broader business goals and thus may resist making investments.
14 Tips for a Successful ITIL Implementation
Tips on completing a successful IT Infrastructure Library (ITIL) initiative.
One example is the IT Infrastructure Library. Back in 2009 when I wrote a story about the the tough economy’s impact on ITIL initiatives, IDC analyst Fred Broussard told me that ITIL’s ability to help IT organizations manage services better, work across organizations more smoothly and make fewer errors might not win it many fans among CFOs. “Money not spent in tracking down errors is much softer than reducing the number of servers or software licenses you are buying,” he said.
Broussard and several other analysts I interviewed for the story agreed on the importance of establishing a clear baseline and relevant metrics right from the start. If you don’t get the baseline right, you run the risk of not being able to demonstrate the kinds of quick wins that will encourage organizations to continue with ITIL.
That point was reinforced when I spoke with Chris Pfauser, a principal consultant with Compass Management Consulting, who told me some Compass clients “do a high-level baseline across all of IT so they can see which areas are really out of whack, then they decide which [ITIL] processes will help bring some of those issues more into line with leading-performer companies, and then focus on those. They measure off the baseline as they are moving forward and improving.”
Pfauser also told me the latest version of ITIL makes it easier to map processes to metrics and provides suggested key performance indicators for measuring success in implementing and managing the processes. As ITIL matures, he said, Compass is seeing “more creative KPIs,” often focused on cost, quality and productivity. He said:
We look at things like the cost per problem and measuring it to make sure it is reducing over time. Your ability to handle more using better tools and processes should mean the number of incidents per person is going up. Change will bring new ones in. But you’ll want to see those going down over time as well.
IT organizations should also involve the business in determining metrics, he said. While the framework’s base metrics work well in the beginning, organizations can often create “tailored metrics that help IT organizations do a better job of reporting their successes.”
Forrester Research analyst Stephen Mann, who tends to offer a refreshingly honest take on ITIL in his blog posts about it, recently shared some great IT service management metrics tips, gleaned from a presentation by Malcolm Fry, author of “ITIL Lite.” Mann also offered his own list of a half-dozen reasons why IT organizations sometimes struggle with metrics. It’s a good list, but the one I found most interesting was: IT organizations commonly see metrics as an output in their own right rather than as an input into business conversations about services or improvement activity.
This hearkens back to another comment Pfauser made in our interview. Business people don’t care about the internal metrics used to measure ITIL, he told me. Instead, he said, “they want to know the business results and how it’s going to affect the cost and quality of the services they get.”
That ties right in with the big takeaway from Fry’s presentation, a suggestion for IT organizations to take lower-level ITIL metrics and use them to create a single relevant view for the business. You should read Mann’s post for Fry’s entire list of IT service management metrics tips, but here are my favorites:
- Collect everything but only report that what needs reporting
- Metrics will evolve as trends and futures appear
- Don’t set targets that wrongly drive behavior