I went to work at a startup after an MSc. I was there for little over a year, and now I am back in Grad school (the same where I was, in the UK).<p>Most of the differences you describe I experienced as well, except that holidays could be planned at a much shorter notice at our company, even though workdays were considerably longer (10 hours).<p>I also had to argue for, and then introduce, automated testing, and some other good practices that I'd expected to be industry standard (was nevertheless nice there was room for this, and it was appreciated in the end...).<p>After a while, though, I missed the research, the self-development, the thinking about broader problems, and also, even though the team was great; the social side of things (especially as the hours left me exhausted in the evenings and even into the weekend). I guess I'm someone who is happiest and most creative when working 8 hours or so, and not tired. I also missed my friends in Europe, and the life-style there (walkable cities, no abject poverty, etc).<p>As for working on important problems, yes I loved what the product tried to accomplish, and our company mission, but 80-90% of most peoples time was spent on bug-fixing, repaying/circumventing technical debt, and at best relatively minor features that most users probably would/should never notice. So on balance spending 10-20% of ones time on a new real-world problem, versus, say 70-100% on a theoretical one, might still end up in favor of the latter as for making a difference...<p>Everybody should make their own decision on grad-school vs industry, but trying both is definitely a good idea.