You’re really talking about two different things here, but you’re wrong on both of them. For starters, your typical software engineer hire will cost you more t

You’re really talking about two different things here, but you’re wrong on both of them. For starters, your typical software engineer hire will cost you more than the value they provide for at least their first six months. If they stay for less than 18 months, then you’ll never get a return on that investment. Realistically you need them to stay longer. Secondly, even if your just a vanilla web dev shop, using the most basic set of frameworks available, if the average tenure of your staff is too low, then the quality of your product/service will massively suffer. Ask anybody who’s had the analyze this in a management position (like I have for instance, many times). To make this work (and I have seen it work), you need an incredibly atypical set of circumstances. For starters, you need to be in a position where quality absolutely does not matter at all. This sounds ridiculous but there are some cases where this can be true (for some period of time at least, but maybe you don’t care about it lasting forever...). Companies that have built themselves an unassailable position in the procurement pipelines of government or large enterprise can exploit such dynamics, on a limited term basis, for example. Secondly, you need to have eliminated any quality control in your recruitment pipeline. Naturally, this will result in hiring a lot of very low quality labor, but as above, maybe that’s fine. Finally, you need to have completely eliminated any onboarding or training. You need people to turn up to work, and almost immediately start producing low quality work. Such companies do actually exist, but they are very, very uncommon. If you happen to find yourself working for one, though you may be underpaid and treated poorly, you can at least know that your employer is likely extracting very little value from you at all, regardless of how personally talented you might be.