As with many people thesedays, I am "graded" at work based on the ever-ominous "scorecard". I've been doing this job without the 'benefit' of one for a while but can't hold it off any longer and have been tasked with basically creating my own scorecard and goals to be held to.
There are two categories that I've put on my scorecard that I could fill out using my own historical performance but my boss would like some other input as well.
The first is "Model Accuracy" -- basically what percentage of reliability do you hold your models to? Do you consider results 90% in-line with reality to be "good enough" or do you aim for 95%, 97%, even Six Sigma levels?
The second is "Project Velocity" -- how long it takes to complete a model. I know this highly depends on the complexity of the model, but any feedback would be great. If I said 90 days to complete a model (bearing in mind I'm usually working on 2-3 at once) would you find that totally unacceptable for any degree of complexity?
Appreciate the input.