I often have career discussions with entrepreneurs, both young and more mature, about whether to join a company. I usually pull the old trick of answering their question with a question: "Is it time for you to learn or to earn?" Let's face it, if you're thinking about joining a startup that has already raised, say, $5 million (as the director of marketing, or as product management manager, senior architect, international business development lead, etc.), the chances of making your retirement money there is extremely small. That's OK. Not every job you have is supposed to be your big break. It's OK for that to be a job where you learn. Yet people often ask me whether I think a company is going to be a big hit. It's clear that they’re confusing learn with earn. So here’s a simple calculation I do for them: OK, you would own 0.25% of the stock. They raised $5 million in their B round. Let's assume that the company raised it at a normal VC valuation and gave up 33% of the company and thus $5 million / 33% = $15 million post­money valuation. If you never raise another round of venture capital (a big if), and if your company is sold for the normal venture exit ($50 million on average for the 200 or so that get sold annually), then what is your stake? $125,000? Yup. Simple math would have answered that, but people rarely do the calculations or think about them. Let's say that it took four years to exit – that's $31,250/year. Now… these are stock options, not restricted stock, so you'll likely be taxed at a short­term capital gains rate. In my state (California), that averages around 42.5%. So after tax you'd make an extra $18,000/year; and that's in a positive scenario! Also, this ignores liquidation preferences, which actually means you'll earn less. Now let's go crazy. Say you get 1%, you sell for $150 million, and it's in three years (e.g., you won the lottery). That's an after­tax gain of $287,500/year for two years. Not bad. But wait a second… Doh! Stock vests over four years. You didn't get acceleration on a change of control? Sorry, we'll have to either cut your earnings in half, to $143,750, or you'll have to complete two more years at whichever BigCo that bought you to earn it all. Either way that money’s earned over four years – so it's $143,750/year for four years. Don't get me wrong. This isn't shabby money. Most people would love to make that much in four years. But don’t confuse getting stock in a company with retirement. Given that a decent home in an expensive area such as Palo Alto or Santa Monica will set you back $2 million, it's hardly riding off into the sunset. It’s why Jason says “the risk­adjusted economics of starting a company suck.” I'm not trying to depress you; I'm just trying to be realistic. If you want to earn – and by earn, I mean the chance to buy your house outright – you have to start a company or join as a senior executive. Or you have to hit the lottery and be an early middle management player at Google, Facebook, MySpace or Twitter. Let's be honest: how many of those are created per year in the entire country? One? Two, max? I spoke with an investor recently who told me that 1,500 deals get funded every year in the US; 80 (5.3%) eventually sell for $50 million, and only eight (0.5%) eventually sell for $150 million or more.