Defining Dashboards
Dashboard are central to agility. They enable us to publish our status in a meaningful and concise way, so that we can avoid having status meetings. Dashboards also provide a way for people to work asynchronously, which is essential for global teams.
For dashboards to be effective, there needs to be a dashboard culture. That means that people feel obligated to create and maintain effective dashboards, and they also feel obligated to check the dashboards of activities that affect them. See Creating a Dashboard Culture in the panel at the right.
A dashboard should have both leading and trailing metrics. Trailing metrics measure outcomes: they tell you how well things are actually working. Leading metrics measure behaviors or other indicators of intermediate progress. These are the behaviors that you believe will lead to good outcomes, and the indicators that will predict good outcomes, respectively.
For example, product failure rate in product or actual use is a trailing metric: it is an actual outcome. Product test coverage is a leading metric: it indicates what the product quality is likely to be, and it reflects your belief that more testing will improve quality. As such, it helps you to adjust behavior, such as investing more effort in the creation of tests.
The metrics should always include the one metric that mattes. (See the panel at the right) There might be more than one, but the point is that at any time, there are a small number of metrics—maybe only one—that tells you how close you are to the goals for the capability that you are working on. Once you achieve that capability, the one metric that matters shifts to a different metric. Thus, your dashboard might change over time, always focusing most prominently on the most key metrics that indicate how well things are going.
Thus, a dashboard should not be cluttered like a static airline cockpit. A dashboard should show the most important indicators that tell you about your progress.
Trailing metrics should be chosen that,
Show the performance of each component – each technical capability.
Show the performance of the entire system – the capabilities that the user is interested in.
Leading metrics should be chosen that indicate the use of behaviors or practices that you believe will support the improvement of the trailing metrics.
All of these should be visible in a dashboard. In most cases, the dashboard will need to be updated manually: keeping it current is extremely important if the organization is to have a culture of using dashboards instead of status meetings.