"Do you have any idea the amount of effort it takes to train 36 people?" It takes a long time for new people to have a real understanding of the current system

"Do you have any idea the amount of effort it takes to train 36 people?" It takes a long time for new people to have a real understanding of the current system, the tradeoff decisions and culture that led to the current system. And it takes a long time for new people to learn how to work effectively together. I'd think bringing in new leadership and product vision first, and having the existing team be given the go ahead to work towards those new goals should be the first step, only doing large scale replacements if they don't deliver. Loads of people will point to the Spolsky "iterate, don't start from scratch" story about netscape. Replacing large parts of your engineering team vs just improving the team you have seems like the same "mistake". Also not sure how they have such a large team in the first place, but that may be part of the root reason. I don't just mean large engineering team - google is showing ~200 employees altogether, which just seems like a lot, but many of those might be CSR positions? But I suspect this is all to do with bringing in much cheaper people.