Velocity is how much work a team can handle in an iteration. It's largely estimated by looking at previous iterations, and it's important because it gives you an idea of how much work a team should be able to take on in future iterations.
Almost every Scrum team tracks velocity, whether on an Excel sheet or directly in a project management tool. Pivotal Tracker offers a nice way to see velocity directly from your project list.
Velocity on its own, however, is not sufficient in assessing if a development team is improving or not. In order for velocity to serve as a meaningful gauge, you need to know how much it varies from one iteration to the next. It's also very useful in seeing how planned work measures up against completed work.
Volatility is a measure of how consistent your team’s work output is. It shows how stable your velocity is, iteration after iteration.
The reason you should strive to deliver a consistent amount of work every iteration is that it makes projects more predictable. Ultimately it helps you make more confident projections.
Favorable volatility is also an indicator of how mature your team's process is. It shows that almost no external force can affect an ongoing iteration. It's also a strong indicator that your team's estimates are on point and that your team is mature enough to overcome, or at least plan for, impediments.
So what does good volatility look like? It's actually pretty boring, as it's a flat line showing 0 all the time. Things only get interesting when the velocity varies from one iteration to the other.
That number is based on the velocity of the previous 4 iterations. If the percentage is going up, it's time to start asking questions.
2. Planned velocity
Comparing velocity to what was planned is a good way to improve estimates.
If your team is consistently delivering more than was planned, it's a good sign you can plan to add more work. On the other hand, if you're consistently missing the mark, it could be a sign that you need to lower your expectations.
I personally love to look at this over time and watch that gap shrink. The closer your team is to the achieving the planned work, the more stable and mature your process is and the more predictable your projects become.
More filtering flexibility
Every team has it's own process and needs. Some teams work across different projects and they split their work into sub-projects using labels. We made Insight highly flexible to accommodate the most complex of scenarios. To do so, we allow the velocity chart (and most other charts) to combine multiple projects and allow for filtering by label, people, and time frame.
For example, imagine you had a label for a new billing module you're building. Now imagine that module is split across different projects and teams. With this Insight's filtering, you'd be able to track the velocity for that module across all your projects and teams simply by selecting the desired labels and projects.
Tracking volatility to ensure consistency of work output and compare actual work against planned work is a powerful way to asses the maturity of your team and the consistency of it's work output. It is also a very good way to have a more realistic projection of your work.