Metrics are Application Objects that serve to aggregate business measures into a certain business context.
The following is a quick reference table for this MicroStrategy Object:

The following diagram represents the logical definition of a Metric:

Where a Metric is the result of a function f (that can be SUM, COUNT, AVG, and so on) applied to a Fact (that can be any raw business measure, like revenue) that is aggregated to a level represented by an Attribute (that can be any business context or business dimension).
The main difference between Facts and Metrics is that the former is a raw business measure whereas the second is aggregated using a function such as SUM, AVERAGE, and so on, and it is associated to a level or context (an Attribute/hierarchy).
The function can also be applied to an Attribute to generate count-type Metrics.
...The function can also be applied to an Attribute to generate count-type Metrics.