[marssaxman]

Startup stock is truly imaginary, yes, but even Google stock has the magical property that some amount of it stops existing when you leave, which everyone does eventually.

...unless they've stopped issuing stock with a vesting schedule attached since I was there, of course, but that seems improbable given the financial advantage it creates for them.

[jholman]

Okay, the stock in the 4-year stock grant is imaginary in the same sense that your future salary is imaginary. I mean, fine, call it imaginary if you want, but that's true of all pay.

The Google vesting schedule, AFAIK, is 1-year cliff and then quarterly vesting thereafter. Which means that, past the first year, at most 90 days of stock is money that might vanish if you leave. I agree it'd be a lot nicer if it vested daily, but it's completely silly to characterize this as imaginary. It would also be completely silly, supposing that you'd worked at Google for 17.9 months and gotten 15 months of stock vestiture, to describe the stock you get 3 days later as a "windfall".

The point of all this is that it is completely reasonable, when discussing how much Google paid a person, to sum up their base pay, plus their bonus, plus the value of the amount of stock that they vest in a year.

If you're choosing between an offer at Google, say with $110k plus 15% bonus target plus 4 years of stock currently valued at $180k (i.e. roughly $45k per annum, probably more by the time it vests), vs working at a startup offering a base pay of $150k (plus say 10% bonus target dependent on revene, and some joke amount of stock valued at some joke number), I think it's pretty obvious that purely from an expected-value-financial-benefit standpoint, you should go with El Goog. Because "total comp" is a real thing. (Not that any sane person thinks that's the most important aspect of such a choice.)

[marssaxman]

Look: I ran those numbers, I made that choice, and going to work for Google was one of the few career decisions I can clearly call a "mistake". The stock was almost entirely imaginary for me because I didn't stay long enough, and I didn't stay because I had no other way to get out of the pointless, demoralizing situation I was stuck in. So, who cares what it could potentially have been worth if the world had been different? The world is what it is, and the money was, in fact, nonexistent.

This has been true of "total compensation" across my whole career: base salary is real, but stock has never amounted to anything significant. The companies whose stock is measurably worth something have turned out to be terrible places to work, and none of the startups have panned out. (Real Networks managed to be both, during the dot com bubble; the expected future value of my options peaked somewhere north of $1.5 million, before the crash, but I ultimately cleared $0. And was so glad when I walked away from that miserable experience.) So, I consider the value of stock in a "total compensation" package to be $0, and make my decisions now on base salary alone.

If it so happens in the future that stock "money" becomes real money, well, that'll be great, but I'm going to think of that like winning the lottery. It has no incentive effect.

[kirovsergei]

You didn't even stay 1 year? Honestly, if your normal experience is to leave your company in less than a year, I think you have some expectation issues to address. While I know google isn't the amazing dream land the news makes it out to be (I've worked there in the past), it's really not that bad. If you've quit that quickly -- and this has happened multiple times -- you should really try to re evaluate what it is you're looking for from your career and see why these jobs aren't working out for you.

To get back to the original point, treating the stock 4 year sum on the same scale as the salary doesn't make sense (though recruiters will pitch it like that). A more honest accounting is to look at the annual number, which is directly comparable to the annual salary and will give you the true picture of how much you make per year. 180k of stock really means 45k per year of stock compensation. For a company like google that's more or less equivalent to cash compensation.

[marssaxman]

I got close, but no, I didn't stay quite a full year, and of course that's not normal! I don't think my bad experience at Google was entirely normal, either - certainly not compared to the experiences of friends who worked there before, whose glowing reports were what convinced me to give the place a serious look, or to those of friends who are happily working there still.

And yet - my Google experience was pretty much the same as the experiences I've had at other large companies. The pattern is that I go in thinking that a big company means big reach, big resources, and consequently big opportunities, so it'll be a chance to dig in, learn from the obviously smart and competent people who built whatever it is that made these companies successful, do some serious work, and level up my skill set - only to find myself stuck on some backwater project nobody cares about with managers playing musical chairs and no way to get out until I've slogged my way through some term of drudgery proving that I'm worthy of better things. Well, fuck that: life is short and I'm not here to waste it. Besides, I've never managed to work productively for very long at all without some internal motivation, and trying to slog through pointless work simply so that I can hang on in hopes of doing meaningful work at some point in the future leaves me feeling stifled, frustrated, and ultimately depressed. (Especially when the work in question is full of tedious, inefficient process that has nothing to do with the actual engineering, which was fortunately less the case at Google than elsewhere.)

At this point I can't imagine what it would take to make me consider working for a large company again.

So the risk of accepting compensation in stock appears effectively constant: you can trade the risk that the company won't go anywhere against the risk that you'll be miserable, but you can't count on any significant return.

[QuercusMax]

Google's vesting schedule is actually 1-year cliff then monthly vesting, but there are 4 blackout periods per year when you can't sell your stock (except on a monthly auto-sale program). Aside from that you're spot on.