Are performance reviews broken?

The company I work for conducts quarterly performance reviews. As a member of our small QA team I've been ask to help define this process. A lot of what I do is identifying issues and collaborating with the devs to reach a solution. How should performance review for QA be conducted? Would any hard metric be useful? I fear that if a metric were ‘number of issues identified’ (for example), that might lead to a focus on creating a large number of issues. That scenario would not bring any value. Any suggestions would be appreciated.