(Bearing in mind this is not from the Valley POV, but I recognize a lot of the same dynamics here in Singapore) Psychologically: The founder is stuck between

(Bearing in mind this is not from the Valley POV, but I recognize a lot of the same dynamics here in Singapore) Psychologically: The founder is stuck between a rock and a hard place. With limited cash to work with (seed rounds are tiny here), that extra 10-20% could translate into 10x that in revenue if used on say, Google ads instead of your salary. On the other hand, equity is also scarce; the founder has just given up 6+ months of his life and career and taken enormous risks to start the business, it's starting to take off at last (hence the hiring ads), he got some funding closed after months of negotiation, so he considers that his equity stake was "earned" and a newcomer "series A/B employee", even at early stage, should go earn theirs as well. There's often not much equity left anyway, not enough for control even after a seed round, so founders often don't have that much to give up. More cynically, this often leads amongst less technical founders and management to the "let's hire young, cheap guys, they work harder too, these experienced guys have ridiculous expectations". I've stopped counting the number of times I heard that (sitting on the management side, advising other companies, etc.). Just last quarter, it was a major angel investor who told my client in his advisory meeting "you should get some unpaid interns from universities to do QA, and offer the best ones a paid internship". Just last week, a colleague who wants to start his own thing refused my offer to introduce him to experienced engineers I've enjoyed working with, because he'd rather "go for cheap people who'll work for free initially". (As an aside, this fear is misplaced. I know of at least two CTOs who worked for "free" for 3-6 months until either revenue or funding happened; the second was in his 40s and had 3 kids...) The investors are a significant source of pressure. Not to say they all do this, but a lot of the more "MBA" types consider headcount a KPI to be optimized, so they push you subtly or overtly to reduce salary, hire cheaper people, negotiate harder, and so on. Think of it as reducing the cost of buying out of the money options (from their point of view). One HNWI who I once considered a friend stopped talking to me after I pitched him an idea, he offered me seed funding on the condition I take no salary until revenue, and I declined (having better options). For some reason, that made him very angry... It's considered a key business skill amongst those people to get "cheap" talent and a failure of your management skills if you do not. This is even more the case when your choice of tech stack and management style encourages wild growth in the engineering department (think Rocket Internet companies with dev teams in the hundreds) and where that 25-50% discount on market rate translates in quite substantial savings. Finally, both from investors and founders, comes the idea of "hedging". Why don't a lot of startups use Haskell or Lisps? Not because they can't find talent (there's plenty) but because it's a lot harder negotiating with a Haskeller with 10 years experience and a PhD, even with the "Haskell discount" priced in (they will take less money to work in a more interesting stack) than with a fresh grad from a local university with some basic understanding of RoR or Django and hopes they will become the next Zuckerberg. This makes the latter a lot more replaceable, and investors don't want their companies hostage to the tech talent. The rest is really stuff that is somewhat right. It IS nicer psychologically for certain personalities (mine included) to have more responsibility, less bureaucracy, a friendly and slightly hipster work environment, a younger set of colleagues full of hope and energy, and a fast feedback loop (it's easy to become indispensable and add a lot of value everywhere in a startup). It IS somewhat true that you will gain experience - if only seeing what not to do - that will be very valuable once you start your own thing, and that spending a couple years on a reduced salary hopping a couple of 10-100 employee startup jobs can be a better use of your time as a future founder than doing AI research at Google.