I started my PhD in software engineering after 20+ years of industry experience. It was an interesting experience. For me I was able to delve deeply into the area of model based software engineering. (There is far more to it than just UML) By graduating, I ticked off an important (to me) item on my bucket list.
As I see it, there are two ways to engage in research. You either have an area that truly interests you and you are excited to deep dive into it for many years, to write about it and to give talks about it. The other way is to work within a research group on the principal investigator's area of interest and carve out a small niche within that. The former path (the one I chose) is difficult because you might end up with very little outside support. The latter, as I have observed, is optimal for those who have ambitions to forge ahead with a career in research. From what I have seen, it is the latter that gets the bad rap.
From a financial perspective, a PhD is not a good choice. But there are exceptions to that rule. If you are concerned about $1m opportunity cost, then it is a No Go zone.
As an adventure, as a massive self-growth experience, for cerebral persons I can recommend it.
With a MS you expand you knowledge by learning from others. With a PhD you create new knowledge and stake your claim upon it. It might be very narrow and specific, but it is yours. Well at least until others build upon it. That is how science advances.