Edit this is answered fairly well here: https://www.parkworth.com/blogs/pre-ipo-tech-giants-using-do.... The TLDR is SEC rules and limited perceived upside of options (although I imagine that could be solved via a lower strike price).
For employees at very early companies that are going the venture route, ISOs are a no-brainer. The company is small enough that the strike price isn't too onerous, the company is too small to hit up against the IRS limits, and they provide pretty good tax treatment under the assumption that the company will grow massively in value - like, 100,000x - which is the the optimistic case that everyone wants to optimize for.
For employees at late stage companies (e.g. last funding round before IPO), ISOs are a rough deal. The strike price is large, so the only people who can afford to exercise them before a liquidity event are people who are already independently wealthy. The tax benefits are also still present, but smaller, because the expectation is that the company might grow 10x in valuation, but not 100x or 100,000x (most $100M companies are not going to grow to $10 trillion in valuation).
RSUs avoid that problem, by requiring zero cash up-front, in exchange for less favorable tax treatment in the "company grows 100x-100,000x" case - which is fine, because that's less relevant.
Of course, the billion dollar question is where the inflection point happens - when do RSUs become a better deal than ISOs? There's no universal answer to that, and some of that depends on specifics of the company, and some of that also depends on who you ask (certain people will benefit more than others from the switch at different points, so it depends on how much the company is weighing each of those [metaphorical] stakeholders).
ISOs also have one other advantage for companies: because they have to be exercised within 90 days of departure, a large portion of ISOs that are granted will never actually be exercised (the employee will choose to leave them unexercised, either because they don't have the money to pay for the exercise price + taxes or because they don't want to). So every option granted is <1 share actually given up (in expectation), allowing the company to grant bigger compensation packages (because some portion of those will not actually be used, and can therefore be reallocated to someone else).
With RSUs, every RSU granted is 1 share actually given up (except in the case where the RSUs expire, which makes the company look bad).
The progression is usually:
- Founders get shares with a re-purchase option for vesting. The company exercises the option if you leave before vesting ends to take back your un-vested shares.
- Next ~100 people get ISOs.
- Next ~2000 get NSOs.
- Then, double-trigger RSUs until the company goes public.
- Then, single-trigger RSUs.
I agree with your general assessment. The difference between options and RSUs usually comes down to upside potential. An option is worthless at grant time (by law, it usually has to be issued at the 409(a)) and it's just the right to buy company shares. If you're buying the shares at the current price, there's no value in that. Options only gain intrinsic value of future appreciation in the underlying equity past your grant date.
On the other hand RSUs are shares of the company, so they are worth at grant whatever a share of the company is worth.
An option with a $10 strike price to buy shares of a company whose 409(a) is $11 has an intrinsic value of $1. An RSU of a company whose 409(a) is $11 has an intrinsic value of $11.
Companies generally, in my experience, give you about 3X as many options as they would shares for the same role. Give or take. Companies usually switch from options to RSUs when they think that growth in the stock price is going to slow down - [edit] (and when they're hiring people with a higher aversion to risk!)
Niantic did 10 years as far back as 2015.
The benefit of double-trigger RSUs over single-trigger RSUs is that they're not taxed until after a liquidity event (IPO, acquisition, etc). That's nice for the employee as they don't have to come up with extra cash to pay taxes on the RSUs as they vest but before they can sell them.
However, double-trigger RSUs have to expire within 7 years -- otherwise there's not a "substantial risk of forfeiture" and they'll be taxed immediately upon satisfying the time condition, just like single-trigger RSUs [1]. It makes sense -- there's no practical difference between a double-trigger RSU that never expires and an illiquid single-trigger RSU, so it would be a tax loophole to treat them differently.
[1] https://drsfp.com/insights/pre-ipo-rsus-single-trigger-vs-do...
