How Do Agile Metrics Actually Boost Team Performance?

The management expert and author Peter Drucker famously said, “You cannot improve what you don’t measure.” Eliyahu Goldratt, author of “The Goal”, said, “Tell me how you will measure me, and I will tell you how I will behave.” Both quotes highlight the importance of metrics in everyday business operations.

Why do metrics matter?

If your team has a desire to improve, you have to first find out where the team stands currently. This simple rule applies to many other domains; from business strategy to personal goals, it is nearly impossible to change our current state unless we know what that is. 

If we wish to lose weight and live a healthier life, we need to know how we are doing against the standards; is our cholesterol or blood pressure too high or too low? The only way to know is to measure it. 

Another critical step towards improving your current state, whatever that might be, is to establish a desired future state. We must define what “good” looks like so that we can work towards it, and be able to assess if we are making progress as well as when we reach that goal.

As Goldratt noted, external forces and perceptions heavily influence us. Whether it’s our management, our partners, or our peers, we are ‌sensitive to how they perceive us; how they evaluate our performance has a tremendous impact on what we do. 

Impact of metrics for your team

So, how do these thoughts relate to Agile metrics for your team? 

We need to be careful about how we measure our teams and their performance, because using the incorrect method of measurement can lead to negative consequences. If we are interested in improving team performance, we must first define what the desired end-state looks like. 

What does “good” look like?

Is achieving a 25% increase in story points what we want? Or, is it a 10% decrease in rework? Alternatively, do we want the team to reduce technical debt by 25% before Release 2.0?

Defining specific measures is a critical step that many teams do poorly. If we want more output, perhaps measuring an increase in story points is the right approach. However, we must consider the possibility that more points completed does not actually translate directly to higher value delivered to the customer. Similarly, a decrease in technical debt, usually a positive, may or may not mean that the team has found a predictable, consistent approach to ensure that technical debt does not grow again in the future.

This may feel somewhat backwards, but to ensure you choose the proper metrics, I recommend you first consider the desired behavior, then select a metric associated with that behavior. 

What metrics make sense for you?

If you want the team to improve their ability to plan and forecast work more consistently, one metric you can track is percentage of work completed versus planned. 

This metric is easy for the team to calculate and monitor without doing much extra work; most Agile management tools will provide this by default. Percent complete provides direct insight into the team’s ability to plan and execute work, and also adapt to new learnings during a sprint or program increment. Define success as achieving a stable and repeatable completion rate.

Another example: If you wish to improve the quality of the work produced by your team, defect rate would provide great insight into how the team is performing. There are other related metrics that the team may use to achieve this, such as unit test coverage, which is also relatively easy to establish and monitor.

In summary, applying the right metrics for your team is a tricky challenge that requires some experimentation. Often, it makes sense to encourage the team to come up with their own measures to monitor. If you can help them understand the desired behavior and outcomes, and provide the space for them to define their own metrics, your team will take a big step in the right direction towards becoming a high-performing team.

Eugene Lai, Contributor