In most workplaces, the big question is to identify the set of metrics that best measure a team's performance and impact. And at a larger scale, for the company itself. And on a tangent, some people like me think of how to measure their individual performance. 

I've been in several debates on what's the right way to do this, especially for teams. 

The first mistake that we do is to look for a single set of metrics to carve in stone and measure all the time. 

There are two questions to answer. 

Are we generating the right impact against our goals?

Are we executing as per plan?

To take an example, the goal of a gaming app may be to increase user retention. If we measure only the impact that we drive, in this case improving user retention, then we lose information on our execution capabilities. At the same time, if we only measure inputs, in this case, features shipped or development velocity, then we lose information on the impact of our work. 

The right thing to do, of course, is both. However, while understanding the nuance that each metric answers a different question. 

If our impact metric is good and our input metric is good, then we are doing everything right. We have a great plan and we are executing well against it.

If our impact metric is good and our input metric isn't good, then something else is driving the impact and not us. 

If our impact metric isn't good but our input metric is good, then we need to re-evaluate what we are working on, because we are executing well against a wrong strategy. 

If both input and impact metrics are not good, then we're in trouble. We definitely know that our execution is bad, but we don't yet know if our plan is bad. So, we fix what we know is a problem for sure and then re-evaluate. 

This way, we're always in control of our destiny.