4 ms·
>Of course as a founder & CEO you know that your 18% of the company at an $80 million is worth more than 26% of a $30 million valuation but I promise you that t
by azov 13y ago
>Of course as a founder & CEO you know that your 18% of the company at an $80 million is worth more than 26% of a $30 million valuation but I promise you that this is a nuance that even really smart & educated employees struggle with.
Oh, really?.. Your smart employees suddenly stop comprehending basic math when it comes to their future compensation?
Or maybe - just a guess - when the CEO was recruiting engineers (who, perhaps, don't have much experience with the business side of startups) he told them they would have 0.75% of the company but forgot to mention dilution?.. Could that be the "nuance" employees struggle with?
- michaelochurch 13y agoI think there's a different situation altogether. It's not dilution that pisses the engineers off; it's the general fact that what they get (equity-wise and career-wise) is so inferior to what the startup mythology promises. Dilution itself just becomes a touch-point, because all that other crap they look back on and realize they "should have expected". Most startup engineers think the equity is a signing bonus, almost like a gift-- say, a teaser-- and that they're going to get an order of magnitude more as long as they perform. The fact that it vests over 4 years (with cliff!) and that initial employee allotments are almost never improved is glossed over until they actually join the company.
- timr 13y ago"It's not dilution that pisses the engineers off; it's the general fact that what they get (equity-wise and career-wise) is so inferior to what the startup mythology promises." Oh, please. You're overselling your hand. Programmers aren't victorian textile workers -- they have their pick of jobs, all with fantastic perks, great working conditions and greener grass just around the corner. Nobody is holding a gun to your head and telling you to work at a startup. If you don't think a particular job is a fair deal, then go do something else. Unlike most people, you have lots of great choices. There are many real problems with software as an industry (e.g. age and gender discrimination), but if your worst complaint is that you're not getting as rich as you thought you might from your day job thinking hard thoughts and typing, then you need a perspective adjustment.
- 001sky 13y agothinking hard thoughts and typing... Meh/. This is a weak argument. Most people in nearly every company from the PA to the CEO sits at a desk with a screen and makes clicking sounds 95% of the time. With the minor process of of 'talking' added, this is the basic description of ~all jobs. It helps to critique the argument at a level of resolution that matters.
- ktd 13y ago>Most people in nearly every company from the PA to the CEO sits at a desk with a screen and makes clicking sounds 95% of the time. With the minor process of of 'talking' added, this is the basic description of ~all jobs. Huh? Even if we restrict things to the civilian sector, there are a LOT of jobs that don't involve that-- police, firefighters, EMS, doctors, nurses, farmers, manufacturers, construction, etc.
- 001sky 13y agoEasy enough. Happy to qualify: Jobs relevant to HN. Cheers.
- tks2103 13y agoYou are arguing more with an imagined opinion and worldview than you are arguing with the comment that you replied to. The comment you replied to refers to the perceived difference between what "the startup mythology" promises and the reality that many software engineers at startups are given in hard benefits. Some software engineers at startups work 100-hour weeks because they think they will receive massive compensation in the form of stock. They perceive the stock they have already received to provide a much larger amount of raw wealth than it would if they crunched the numbers. They perceive that their company might provide them with much larger grants of stock if they perform exceedingly well in building the product. Sometimes their bosses guarantee them that they will receive a massive cut of the company, if they just work a little bit harder for a little bit longer. I don't know how common situations similar to the above are, but they do exist. I imagine many folks reading Hacker News have been on one or both sides of the table in that situation (stock-granting and stock-receiving). I believe that is what the comment you responded to is criticizing. Nowhere was there a comparison made to victorian textile workers. Nowhere was there a claim that '(not) getting as rich as you thought' was the worst complaint about the software industry. You made the latter two arguments up, and then knocked them down.
- warcher 13y agoRun your own shop then. I mean, if you don't like how the pie gets split up, bake your own. There has never been a better time in the history of mankind than our industry, right now, to put your money where your mouth is. You don't need an Ivy League degree, a rich daddy, or a million in equipment, just a laptop and a dream. That's.... Almost inconceivably powerful, given how other businesses operate. And I don't think it will last forever, either. I did, and I'm not sorry I did, but I can say with confidence that this shit is not for everybody. AND I paid a heavy financial toll for 'equity', which is nonsense, but I got agency, which is priceless.
- louischatriot 13y agoI don't think it's the job of a startup CEO to explain startup mechanics to his staff.
- jacquesm 13y agoSorry, but it is exactly the job of the CEO to do that. You don't get ahead by keeping those around you in the dark with respect to the rules of the game you are all playing. Doing that consciously is pretty low. What you want is for everybody to have as thorough an understanding of these issues as time and resources allow so that everybody will be on exactly the same page. Only by maximum alignment of your interests will you maximize your chance of success.
- louischatriot 13y agoYou misunderstand me, I'm talking about explaining how your startup works and what the business is. The startup mechanics I'm talking about is the way we generally work in startups, for example the fact that if you work for a post series A startup you have to expect quite some dilution before na exit.