My nicest boss ever, once told me offhandedly "I don't associate you with making money at all". I am sure many of my bosses felt the same way about me. They just didn't say it out loud.
I have always worked in research and development departments, and these are known in the trade as "loss centers" where the money bleeds away, as opposed to "profit centers" like sales, where the money flows in.
Needless to say I take a slightly more balanced view. Cutting away all R & D is a common way of cutting cost and improving the profit-to-loss ratio or the productivity per employee or some such metric of a company, in the short term. But it is a lot like slash and burn farming. It is not something you can do in the same place for a very long time.
I am not much of a businessman, but I have been in the game long enough to pick up some tricks of the trade.
Investing is not necessarily a bad thing, but economically it only makes sense if future profits warrant the initial outlay. The question then becomes : how much profit will justify which investment ? This metric is known as ROI. The acronym stands for "Return On Investment".
It would seem simple enough. If the future profit outweighs the investment ( including some interest rate, and with a risk factor taken into account ), then the investment is worthwhile. But this is not the whole story.
Anyone with money to invest ( or their own time, even ) has a choice on how to use it. At any given time, there are always many opportunities. There is a kind of market here. Some of these opportunities will give a future ROI of $500 for every $100 invested, others maybe twice as much. If the going rate is 10:1, then it is not a good idea to invest in something which will yield only 5:1, because you can invest every dollar only once.
An investment of $100 is "worth" $1000, if you have the $100. The cost of missing out on the $1000 return is called the "opportunity cost". So, a boss does not rate one hour of development work as the $100 investment that the employee and their overhead may cost. They rate it as an opportunity cost investment of $1000.
The opportunity cost argument is hard to swallow for a young and enthusiastic engineer with a great idea.
Instinctively, they will feel that they should get funded as soon as the ROI is positive. It feels like a grave injustice when the greedy boss says that he needs another factor of 10.
Most people, but especially students and university professors, somehow feel that their own time is "free". It has already been paid for by a government grant or by someone else, and the burden of proving a certain return on investment is lifted. The pleasure of working in such a low stress environment ( or in your own garage ) can outweigh earning more money in industry.
I am definitely not saying that a professor of archaeology should have to prove that their science, or culture in general, is useful in some foreseeable future. The future is not foreseeable, and culture has a value unto itself. Not everything can be readily expressed in cash. But even in such environments it can be useful to consider that an hour can be spent only once.
Returning to the bare cost ( not the opportunity cost ) of an hour's development work, there is always another trade-off which is known as the "make or buy" decision.
You can copy a page of text by writing it out in longhand yourself, or by making a photocopy. In your spare time, the choice is yours. You may value your own time at zero, and make your own lunch. But working in a company, or after deciding that your own time is a limited resource too, some trade-offs need to be made.
The amazing thing in universities, and in unprofessional development departments in industry, is that buying hardware is seen as "real" money, and hours are there anyway. It is quite common to have to spend three hours as a supplicant and one hour of management time in meetings to get a $50 hardware budget for a prototype approved. Worse, it can easily be demanded to spend another eight working hours to prove that this particular hardware item is the cheapest or most effective possible, and the $50 will not go wasted.
I have been fortunate enough to work in companies where bosses were teaching me the folly of such ways, urging me to buy a few alternatives, pick the right one, and discard the others within minutes. It is so much more rational, and so much easier on everyone's nerves. Offhand, I would say that a development group should be able to spend petty cash of at least $50 per week per person, and sometimes much more if the tools of the trade happen to be very expensive.
My first boss in the aircraft company laughed in my face when I wanted to cycle to the lab in Amsterdam to deliver the daily flight test tape. He was very kind about it though. He just said "you are worth to me so much more than a taxi driver".
... there is always time to do it over. There is a balance between endless optimization (" analysis paralysis ") and throwing money away on ill considered plans. And people do vary in their propensities.
There is the type of person who will go on "gold plating" forever unless forcefully reminded that it is best to leave well enough alone ("don't make it too nice"). No one should be unnecessarily careless, but well documented and therefore recoverable mistakes are not necessarily a bad sign. "If you never miss a plane, you're spending too much time in airports".
On the other hand, there are bosses who think that driving the schedule hard is the way to keep development cost down. Such people need to be reminded that there is no such thing as "first time right", and the first version of a design is never a viable product, despite exhortations in the software industry to "fail fast", and "if you're not embarrassed by your first product, your first product is late". This may be true of some consumer software, but it is not true of any serious product, particularly in hardware.