Odd - all my perl scripts have not achieved a single dollar in savings. They've merely given me (and others in my team) more time to do things we wouldn't have been able to do otherwise ;-)

My last manager asked me this type of question all the time - estimates of cost (and thus savings) was a significant portion of my job. We had the conversation early on where we both knew that the numbers were all complete lies, but, and here is where some programming purists get flustered, that's just the way the corporate game is played. The rules suck, but, based on the golden rule (he with the gold, rules), if you want to get paid, you gotta work within the structure given (ethical and moral quandaries aside).

I just got into the habit of making stuff up that was plausible enough to pass muster. Sometimes my made-up numbers would widely disagree with my manager's made-up numbers, and we'd have a discussion about why the numbers should be more or less than what we thought they should be. Which may sound strange if I'm also claiming they were all numbers just pulled from out of the air. But that's only part of my claim. The other part is that these numbers form part of a corporate game. That game is one whereby you attempt to quantify your contribution to the company and the toll that the company is taking on you.

It's actually like my old joke node on XP. But more serious. This really is an attempt to take all your contributions and worth as an employee, and boil them all down to a simple, easy-to-grasp number. Easy enough for a non-techie (often higher management and/or HR bean-counters) to grasp.

If you don't like the numbers (which I'm not blaming you for - though I may play the game, this doesn't mean I like it), think of it this way. You're trying to come up with the value that your perl scripts have provided to the company. Think about the time saved, and what would not have been accomplished if you had to continue doing those tasks without the perl code. The value of those additional tasks is directly the value of the perl code. You just have to convert it to a number so that others that may not grasp that correlation may grasp the value anyway.

So, then, what is the value this year compared to last year when you wrote that code? Well, you won't have to write the code again, I hope, so the amount of time spent on the perl code will be a new savings this year. And, as was mentioned, you probably will have more time to use the code this year than last, so there's more savings. Of course, next year will be a problem here - your savings next year should be the same as this year, all else being equal.

But there's the key: you have more experience this year than you did last year. You should be able to take that experience and tweak your code to help more than it already is. To automate more things. Or just to be simpler.

With all that in mind, you should be able to express two numbers to your boss: a) the savings given nothing changing, and b) the estimate of what more you can do to tweak the process to be more efficient. At this point, it may not be the revolution that it was last year. That's ok - small progress is still progress.