This is part of a continuing series with Bob Anderson, IT Operational Excellence, presented by Anne Grybowski. You can read the entire series here.
Let us return to our cookbook example from the previous article. Once you have followed the recipes in your cookbook with you standard processes and process control, how do you gauge the success of your outcome? What is a metric relative to IT operational excellence? According to Bob Anderson, metrics are quantifiable measurements that represent value and achieves goals. In short, you have to know precisely what you want to measure before you try to measure it:
First of all, what area are we going to measure? The metric has to be in alignment with the functional areas that are going to be measured. It must be quantifiable, which means that it has be measurable and the ability to collect some kind of data with respect to that metric so that we can measure our performance against a goal. So those are the first two steps; what is the function and what is the quantifiable metric that represents value or represents excellence? Second, now we need to have a specific goal for each of those metrics.
Critical Success Factor: Setting Goals
Without a goal to measure success against, metrics will do little to no good. Anderson uses the example of wanting to resolve all critical incidents within a set time frame. Let us say the time given for the task is two hours. That time span of two hours, according to Anderson, is your goal. He notes that “it is a quantifiable element that we can measure success or failure relative to, let’s say, infrastructure or application support.” Once goals are defined according to the needs of a specific business and IT organization, we must determine if they can reasonably be achieved. Make sure you have buy-in from both the person(s) setting the performance goals and those who must achieve those goals. Anderson reminds us that, while goals are important, overly aggressive goals, that are not agreed by all parties effected can end up doing more harm than good to the business.
Now, you and your organization must figure out where and how to collect data. As we know, you must measure your actual performance against performance goals, but what exactly do you want to know? Anderson wants you to ask yourself three questions: Where is the data created? Where does the data reside? How do I get to it? Anderson stresses that you must know the answer to each and every one of these questions before proceeding or else you will not be able to use metrics effectively. Think of those three questions as links in a chain. If one link is missing, everything falls apart.
Once you have collected and analyzed your metrics data, Anderson recommends finding a trusted method of storing them. A file or a data base is needed for information to accumulate. This way, everyone can go to one central location to see how things are running. Your data should be stored in 2 pieces, one detailing exactly what you are measuring and the other stating the performance goal to be achieved. Finally, what tool will you use to display this data? Data must be easily accessible and easy to understand. Once you have everything in place, metrics will be a wonderful tool for you on your way to IT operational excellence.
The next article in our series will discuss Real Time Data Repository, the sixth piece of the IT Operational Excellence Puzzle.
Robert Anderson’s Bio:
Bob Anderson has been with Computer Aid, Inc. since 1988. The majority of his 38 years of IT experience have been spent as a senior executive in large IT organizations. In addition to being published, Anderson has been the principal architect of CAI’s process management, event tracking, and resource management tool, “Tracer” and most recently created a free assessment survey to help organizations recognize where and how they can improve their operations. He has also built CAI’s Production Support & Training department. Anderson is also a decorated US Marine. He and his wife have two grown daughters and reside in Boiling Springs, PA.