I see a few ways of calculating 'incremental benefit' -- optimization, and increased usage.
First, we take the cost of an employee to the company (around here, it's $125k, due to salary, benefits, office space, equipment, etc., but you might have different grades of people, etc.) Then we assume ~47 weeks per year (holidays, vacation, sick leave, etc.), and we come out to ~$66.50/hr.
Now, if you were to optimize your program, and shave off (x) seconds per execution, then it's just a simple time savings based on the number of number of times people execute it through the day (y), and the number of people using it (z)
$66.50 * x * y * z * 5 * 47 / 3600
Now, if you consider that either (y) or (x) may increase since last year, we also have to add in for each of those, but instead of (x), we use the overall time savings of the program over the previous methods (t) (in seconds)
$66.50 * t * ( (delta x)*y + (delta x)*(delta y) + x*(delta y) ) * 5 * 47 / 3600
It's that second one where you can justify your work, normally. Some people might argue that it's an already existing program, and you shouldn't count it, but you don't know for sure that it's going to work for every new input thrown at it, and someone's going to have to maintain it should it not work.
In reply to Re: (OT) Boss asks "Incremental savings through Perl Scripts..." what to answer?
by jhourcle
in thread (OT) Boss asks "Incremental savings through Perl Scripts..." what to answer?
by vkon
| For: | Use: | ||
| & | & | ||
| < | < | ||
| > | > | ||
| [ | [ | ||
| ] | ] |