4 ms·
These are tough questions, and I'm not sure I'd trust random message-board participants (like me ;-) to answer them. People pay operations-research consultants
by nostrademons 17y ago
These are tough questions, and I'm not sure I'd trust random message-board participants (like me ;-) to answer them. People pay operations-research consultants hundreds of thousands of dollars for this.
Something to consider: good code is an asset, and it's a compounding asset. The better your programmers are, the more likely that they'll produce code that they can build upon, and build upon, and build upon, all with relatively few bugs, so that you can go faster as time goes on and the product grows. Mediocre programmers often produce code that requires more maintenance and slower development as time goes on. Either way, the effect is exponential and not linear, which makes it hard to compare against linear differences in salary. Probably should run an IRR comparison against your cost of capital and decide based on that.
This, BTW, is why companies like Google try to find the best talent they can, no matter what the cost. Productivity compounds, and not always in obvious ways. In addition to writing good, solid code, top developers also attract other top developers to work with them, and they create a culture where high achievement is expected and you can't just slack off. These are intangibles that are really hard to put a price on.
- mingyeow 17y agoThis is a great comment! This is probably the best articulation of why "good code is an asset, and it's a compounding asset, and the effect is exponential and not linear" - good insight, well articulated "running IRR" - definitely right. I think you can apply different variables to the various categories of work. Summarizing, the main thing, whether outsourcing or in-house is to find people who can write great code. On my end, understanding of the space you are in is very important as well