That leaves you with a question: How do you tell if your team is successful or not?
The Measuring Stick
You probably use metrics for other aspects of your business; perhaps you measure IT efficiency or you use metrics to determine whether you're providing good customer service or not.
DevOps can be measured in much the same way. If you want to know how successful your team is, simply take some measurements and analyze the data. This information can give you insight into your DevOps performance.
Sounds simple enough, right? Collect some data—some of which you're probably recording anyway—and analyze it to determine how successful your DevOps team is.
Unfortunately, it's a little bit more complicated than that. Just like any other field, there are plenty of metrics to measure, but not all of them are necessarily good measures of your DevOps team's performance. The question you have to ask is "which metrics?"
Measuring What Matters
You need to decide which measures you'll monitor to determine "success" for your DevOps team. Success is, of course, always subjective; what counts as success for one DevOps team may not be the same for another in a different company or even when working on a different project.
What you need to do is define success for your team and then choose the measures to monitor those facets of success.
So how do you choose? There are plenty of metrics; it would be easy to choose one and say something like, “the number of bug fixes your team performs determines success.” The more bugs they fix, the better they're performing.
But is that truly what matters to you and your team? Your DevOps metrics must reflect what matters to you. Do you want to spend time fixing bugs? Maybe you shouldn't be focused so much on fixing bugs as you should be on writing great code from the start—and having fewer bugs to fix as a result.
What Makes a Good Metric?
As mentioned, not all metrics are created equal when it comes to measuring your DevOps team's success. Some measurements yield better information than others, and still others are more meaningful for you and your team than others.
So what makes a good metric? The information and insights it provides. Yes, counting the number of bug fixes might tell you how many bugs there were in your program and how quickly your team can fix them. However, it can't tell you if you're writing great code and have few bugs to fix as a result, or if your team is just not finding and fixing bugs that exist.
Better metrics are more precise; they give you unambiguous insight into what your team is doing and can more readily help define your success. If your focus is simply on writing great code with few bugs, then the number of bug fixes isn't a good metric; instead, the number of bugs might indicate the quality of code.
Less Is More
When it comes to metrics, it can be tempting to simply measure everything; after all, you can probably collect the data without much effort. In many cases, you're already collecting the data! But interpretation is the difficult and time-consuming task.
Just because you can collect the data doesn't mean you should; it isn't necessarily a good metric. And even if you already have the data, it may not be a good measure. Instead, focus on only a few meaningful metrics to determine DevOps success.