1) I really like this, for transparency, clarity, and design.
2) The actual numbers there are a little bit scary. 0.03% equity grant? That's tiny. It's basically worthless.
When did startup equity get so tiny? I'm not singling eShares out here -- just broadly speaking. That means that if eShares exits past a billion dollars, those shares won't even buy a house in the area....
Non-technical line employee past Series A. See also the top 25% salary of $64.5k, which is roughly half of what new college grads get offered in engineering at AppAmaGooFaceSoft.
Fundamentally this is driven by one number which is an iron law in the Valley: the employee options pool gets 20% of the company. This is 2000 basis points. It will never, under any circumstances, suddenly have more than 2000 basis points in it. That's 2000 basis points you can issue to every employee from engineer #1 to social media marketing intern #6 in year 8.
What's the average equity allocation going to be when you have 400 employees? Well, it can't be 6 basis points, because 6 * 400 is greater than 2000. Is 400 employees a reasonable amount of a company at IPO? It's actually on the low side, particularly as companies are waiting longer and longer to IPO.
This means that, as companies get de-risked by succeeding investment rounds, equity available to employees drops sharply. Engineer #1 might get 200 basis points out of that pool, for giving up a sleepwalk-to-$300k offer from Google to take a flyer on a company paying $80k which needs to, ahem, build a CRUD app. (Not casting aspersions on any company by calling them CRUD apps. CRUD apps make the world go round.) That's still a low number relative to the amount of risk that engineer is taking and their opportunity cost.
I've heard on the grapevine that the iron law might be relaxed. Investors are pretty in favor of this, as long as the extra basis points come out of founder pockets rather than investor pockets.
Great answer. I'll just add an anecdote where I was in a fairly similar role/situation as this fictional example of Robert.
The equity grant # was about #350 at a post Series B startup that would exit for about $1B later that year. My stake in the company was 1 basis point (0.01%).
Equity grants loosely follow an exponentially decaying pattern.
So your 1 bp grant was basically a year's salary for a junior engineer in the Bay Area. What is the advantage of joining such a large post Series B startup (which could still face a down round or layoffs before the exit) instead of AppAmaGooFaceSoft?
There isn't one. If you want to join a startup and do well you have to be willing to join when they're paying lower than market salary, desperate for help and nobody's more than 50% sure they'll be around in 2 years. If you can't handle that then a startup isn't a great proposition for you.
It's probably a lot less than even that because, having joined less than a year before exit, the strike price would be far from nothing.
Having said that, you can only say there's no advantage in retrospect. If it had become a $10B or $100B company, 1bp would be substantial. And if the exit was a stock swap then perhaps the potential is still there.
Don't forget those options aren't just lottery tickets; unless your employer is giving out options with 10-year exercise windows, they're rigged lottery tickets. viz the CTOs on here talking about using them as retention leverage.
I assume Zach got interested because he discovered after he was fired from github that he was close to losing all the money he "made" on options when github became a unicorn -- he probably had 90 days to exercise a very high priced illiquid security with horrid tax consequences if he fucked up. That's the thing about cash: it's not rigged. If you get cash, it stays in your bank account. Paper gains can disappear.
How many years until the record setting number of C.S. grads start revealing the shocking truth that they aren't actually getting thrown six figure entry level salaries which sensibly reflects the relatively low number of hires FaceGooBoxSoft make relative to the population of educated western developers?
That's true. A lot of people aren't getting thrown six figure entry level salaries and perhaps won't be able to find one, modulo geography, skill, personal preferences, etc. And it looks like we're seeing the market soften, so perhaps soon they won't be that easy to find.
But as it stands right now, if you:
a) Are in SF or NYC
b) Have a CS degree from a well-reputed university.
c) Are working as a programmer.
100 k$/yr is not only achievable but practically the minimum.
That said, the linked offer isn't for an engineer, it's for a financial analyst, so perhaps 65 k$/yr is in fact market rate or better.
What gives you the idea that those qualities make anyone actually worth 100+ k$/yr?
The only factor in that short list actually providing any indication of worth to the company is "has a CS degree from a well-reputed university" and that still doesn't indicate being able to perform in a job (rather than study).
I wish our economy could move beyond grading employees by degrees. They're proxies for the things that are hard to measure. Sadly I don't think they can easily be replaced (if at all).
Point c is interesting and very insightful (whether it was intentional or not). In order to qualify for a $100k/yr position you must already have a job. This by definition would exclude true entry level candidates, though entry level candidates today seem to be required to have significant work experience as a form of validation.
Definitely agree that the iron law needs to be relaxed. You can give special voting rights to founders stock to make sure they maintain control of the company, but currently the risk/reward ratio between founders and early employees is way out of whack. Glad to hear investors are keen on this since I think it should make startups stronger.
Equity is a key component of modern compensation with better tax implications and more upside.[1] The “built in the garage” myth needs to die and doesn’t justify 100-1000x payouts when early employees are taking a significant risk too in opportunity cost. The main issue currently is information asymmetry as most early employees don’t see the cap table, don’t realize how much others own, don't understand how little risk most founders take, and don't know how much they are truly giving up in total comp from more established companies.
Relaxing that 'iron law' would be a significant blow against capitalism itself. That's what Occupy* wants, and not what VCs want. But VCs write the rules and there's no real pressure to write them differently.
I wonder, if you do the math on equity compensation versus compensation in the job market as a whole, if returns from equity aren't just a tiny, insignificant sliver of overall compensation that we happen to take seriously because of availability bias.
Either way, I do not believe it to be a "key component" of "modern compensation". Tech companies can pay strong salaries just like other companies. They choose not to. That could change, and I think should.
Hmm I was actually thinking 0.03% was high for the role and position of the company.
Startup equity has always been tiny. Even for critical engineer hires 1-10. Engineers in general are poor negotiators and, in my experience, the employee pool is preserved for later executive hires that are brought in to bring the company and product into the fold of the industry trying to be disrupted.
Simple question: Why not retire whole concept of stock options? Why not just give RSUs? More I read about how stock options work and how it can ruin you financially, I just don't think this whole thing is friendly to attract talent. I would be super hesitant to go to any company offering stock options instead of RSUs - no matter how good the offer looks on paper. Even though RSU value can go down to zero, you at least don't end up paying our of pocket. And things don't require 3 pages of explanations and caveats in your offer letter.
A (somewhat) secret: lots of employees don't/can't exercise, returning options to the pool.
Hell, just last week someone was on HN discussing walking away from $0.5m because he couldn't afford to exercise when he left an employer that was later sold.
At the required decision time, the employee often has no idea how much or when a company will be worth money. Plus the obvious risk that someone fucks up and the company is worth a lot less than planned.
So he didn't really "walk away from $0.5m". He walked away from a piece of paper that may or may not be worth $0.5m in some unknown time (or never) in the future.
The ruining you financially is mostly on tax basis, and an RSU forces you to pay tax immediately on vesting as income, whereas an option lets you wait and choose when/how to pay taxes, and has some tax benefits depending.
Which means the employer needs, in cash, enough money set aside to cover the liability of ~35% of fmv(not strike) of all rsu issued. This really is unlikely to happen at an early company.
What you really want, if you join at a point when you can afford to buy it, is to get options and buy and file 83b so you have zero tax owed.
Often when later stage companies start offering RSU there is a double trigger vesting. One is the normal vesting schedule where you earn the stock and the second is when the stock becomes liquid (acquisition, IPO).
This way you wont get the tax hit until the stock is actually liquid. The downside is that you cannot get longterm capital gains tax treatment, instead it's taxed as ordinary income with your income tax bracket.
Personally, I think all the tax optimizing on stock options doesn't seem worth it when most startup stock never see any returns. I wish IRS could find a way to fix this. It seems silly to pay taxes just because you own stock that is not liquid on any market and which you often can't even legally sell without company's permission.
Whats frustrating is ISO fixes this exact situation, but AMT rules haven't really caught up with the current situation(they've barely changed since introduced iirc) - all they have to do is raise AMT to something more reasonable, or exclude ISO from AMT
the real problem is that options are meant to be given with a strike price so low they are basically given to you like RSUs, like 1 cent. Instead companies make the price much higher for no good reason other than to force you to be unable to execute it until the company goes public. If you have to sink a large amount of money to buy the shares and then are unable to sell them to pay the taxes most people wont exercise the option and will wait it out until the company goes public. People refer to this as the golden handcuff that prevents you from leaving the company because you lose your options 90 days after you leave the company. They then take your shares that should be rightfully yours as you worked there for years and give them to the next new hire. It is purely for the company to save money. Any company that has a strike price greater than 1 dollar is being greedy and it should be a red flag that they dont really want you to have those shares
You're missing the role of the IRS and tax law here.
In order to grant ISOs (Incentive Stock Options), which are taxed less unfavorably than NSOs (Non-qualified Stock Options), the strike price of the options must be equal to or above the "Fair Market Value" of the underlying stock.
The "Fair Market Value" is usually determined by an independent valuation firm through a process called a "409A Valuation." Firms like SVB, Alvarez & Marsal, Capshare and others do this.
The later-stage the company is, and the closer to IPO, the higher the 409A valuation will be.
If a company were to grant below the fair market value, there would be several bad consequences:
1 - You would owe taxes on vesting -- this means you could owe taxes possibly as soon as the first day of work, and likely on your 1 year anniversary of work. You owe these taxes whether you exercise or not;
2 - The company would have to withhold income taxes on your exercise date, including "employer-side" payroll taxes.
Normally, with an ISO exercise, you may face "AMT" (Alternative Minimum Tax) but you might not. By contrast, you certainly owe taxes with cheap NSOs.
Companies have to set the strike price based on the value of the company, which they must file with the IRS, and be able to defend(called a 409a). If youre pulling revenue you can't claim the company is worth no money...
Stock options are almost university priced based on a 409A valuation of companies "fair market value" which is a legal requirement. If anything various tricks are played to somewhat lower the valuation # spit out by these valuations.
I don't like the language of "below is what our lawyers require we tell you". The lawyers didn't decide that the offer is contingent on a background check, or that the employment is at will. You decided that, the government insisted that you communicated it to your prospective employee, and the lawyers informed you of that requirement. Maybe it's splitting hairs, but it would leave a better taste in my mouth if the company is took responsibility for those decisions.
Other than that, the letter is great and shows a wonderful commitment to transparency.
Agreed. I personally would prefer if offers had, as a standard, 2 weeks or 4 weeks severance. If the company is not hiring a high percentage of people it needs to fire or lay off, it's not a problem. This of course provides more security to their employees.
All the tech companies are talking about a shortage of skilled workers, but outside of management the terms of employment are still biased toward minimizing the company's potential costs. If there were a real shortage we'd see contractual changes.
Yep, that line is pretty lame, especially if the letter is signed by the CEO as seems to be the case here. Either they are not being straightforward or somehow don't have control of the legal folks. There are various reasons for the latter case, some of them potentially quite negative.
To be fair this letter seems to be making an effort to be transparent, so I get what they are trying to do. As I noted in another comment though this is a tough trick to pull off successfully without seeming insincere.
Nitpick: The slide, "How we set your salary", shows a (presumably) normal distribution (with specified mean and std. dev.) and then uses this to determine the percentile value. (Since a normal distribution is fully specified by mean and variance)
However, is a normal distribution appropriate for income/wages? (Or better: Would this even have an effect on the percentile value?)
It also comes off a little entitled of the company. So what if you have some distribution? Maybe I don't agree with you about where I fall in the distribution.
A policy to set salaries without considering how far above average the candidate is is unsettling. It speaks of seeing people as fungible cogs. If one person won't accept our distribution-based-standard-offer, oh well, next...
As a candidate I would see that and then begin to question whether other good people turned them down, and what caliber is the team I'd actually join.
It's graded vesting, I've always seen it at 5 years (20%/year). It seems they do 20% after the first year then 1/48th afterward monthly (which is nice).
I actually saw this article and was immediately thinking of an additional product my company could sell (we sell recruiting software).
But then I started playing devils advocate. If your sending an offer letter wouldn't you want it really personal and not as much about the $$$. Wouldn't you want to cite memorable and salient specifics about the individual.
This letter seemed like it was completely automated. It didn't seem "sticky" at all. I think a "personal" letter might have greater success particularly for smaller companies hiring expensive talent. Obviously you need to state what the candidate would be receiving for compensation but I think making it the only focus will lead to price comparisons.
Recruiter here. You raise a good point: I think this takes the need for compensation transparency to an extreme.
However, I strongly support this concept overall. Often, I see the offer letter essentially presented as a formality once the basics are agreed upon. The candidate has completed the interview process, decided his/her non-financial interests in the opportunity, and received that "personal" positive reinforcement from the hiring manager & team. At this point: sure, the offer letter can include a personalized message to start, but ultimately it's for the candidate to review the hard facts: compensation, benefits, disclosures, etc.
I applaud this, but I'm cynical and think it would require an overhaul of the entire system regarding compensation transparency. The status quo offer letter simply benefits the company, and it's hard to justify changing course when only a tiny minority of companies are following suit. If this trend took off, then the threat of a competitive disadvantage come offer time would be a worthwhile motivator. I hope it happens someday.
I disagree, not only does this give me a huge amount of insight into the company (more than I've ever seen from an offer letter), if it is automated that's actually a plus to me: it shows that they know this stuff is an important part of the offer and are showing it to everyone. I don't need to hear nice stuff about myself, I need to deeply understand the deal I'm making.
Plus, there's nothing that stops them from adding a personal cover letter to this.
It does feel really robotic. I appreciate the transparency, but it also implies "you're an equation - and as soon as your numbers don't make sense for our bottom line, you're position shall be liquidated"
Yeah, I think a hybrid approach might benefit. Even a section in the middle that is evidently 'personalized' or custom written (even if you guide management as to what to write) might help.
"We are excited by what you have to offer in the data science field - we are looking forward to seeing how you can ..."
But in this same letter they encourage you to not do an expected value calculation about the value of your compensation package. That's extremely bad advice.
The more I read this letter, the more it looks like they are trying to be selectively transparent in ways that don't actually cause them to give up any information advantage over the candidate -- and they are putting overly specific numbers in (like the magic 77MM figure for an exit) and tossing in relatively meaningless graphs, and an incredibly meaningless salary distribution, to add window dressing and make it sound like it's all much more technical, facts-based, and non-negotiable than it really is.
While it is a move in the direction of transparency, which is good in theory, this particular offer letter makes me feel like it's just a wolf in sheep's clothing -- still fundamentally trying to get a candidate to agree to be priced at a level well below their actual value to the company.
If the letter said "live your life as through your equity is worthless, because it is worthless and by agreeing to work for us for a lower wage you're not likely to recover those lost earnings down the road" .. then I might be impressed by the honesty.
As someone who has worked remotely in various percentages, being in the office one day a week can be quite helpful for developing social capital, team building, keeping in sync, etc. Some things, even with video, are just hard to do remotely.
I'd largely agree with that - but that's not what's going on here. It seems that the person entertaining the offer would work in an office in Soma, then commute another 50 miles one way, once weekly.
Honestly, that doesn't seem particularly horrible. Lots of people spend a lot of time driving to spread-out customer sites and/or travelling by air to various places. I know Bay area traffic etc. but one day a week in an office that's a bit of a drive doesn't seem a particularly onerous imposition.
We have two offices, one in Palo Alto, one in SOMA. Our product teams are based in Palo Alto, but client services, customer success, etc are based in SOMA. We want a lot of cross-pollination between those groups and our product teams, so we commute to the main office once a week. It works pretty well – sometimes talking face to face is more effective than through Hangouts or Slack.
I did the opposite for about a year, PA to SF once a week. In our company, it was pretty forgiving. Most people going to SF were in the office from 10:30-4. I'd leave home and be back at roughly the same times as when I went to work down the street.
Water cooler talk & f2f meetings are essential to building intra- and inter-office relations. However, given a choice between a job that mandated this and one that didn't, I'd choose the one that didn't. Given that I have a 25 minute walk to work every day, I'm sure that my worldview is terribly skewed, and I acknowledge how lucky I am to have such a great commuting situation. But that still strikes me as odd. Are product teams under the same mandate?
The capital structure graph is wonderful. Very clear. Would be nice to also send out updated charts to existing employees every time the capital structure changes. All startups and VC-funded companies should do this!
The part about exit values is a bit misleading. You don't include anything about liquidity preference or anything else that would be taken off the top before common shares are bought out.
And if shares are diluted or liquidity preference is added later after some fundraising, do you always go back and share all of the specifics of all of that with all employees? Kudos if you do, but if not it kinda renders the initial transparency less impressive or useful.
I've generally found the advice here [0] to be a much better proxy for hypothetical exit calculations.
One other thought: isn't it odd that you're willing to provide a distribution over salaries to argue for the salary you offered, but you just pull a number out of thin air for the possible exit?
Why not provide a distribution on exit values (and evidence for why you believe it), and then show the expected value along with risk, under that distribution.
At the very least, this way if your distribution over exit values isn't a severe power law drop off with uncomfortably low expected value, the candidate knows you're blowing smoke and can walk away.
Instead, you deny the premise that an expected value calculation is useful towards the end. Wow, that is absolutely walk-away-immediately stuff right there. "The distribution over outcomes is unknown..." Right, so of course you should make zero attempt to gather data about it and model it. That's one of the most bananas things I've ever seen, and it's right in the offer letter!
Picking a number like 77MM is almost like a dirty in-app purchase priming trick or something. (And no, it's not the least bit redeeming to add a tiny line at the bottom of the chart stating that it's just an example and not a forecast.)
If you personally, actually believe no one should engage in an expected value calculation regarding their start-up compensation package, then why not simply omit any slides about what the options will be worth? The 77MM thing is a red herring, and surely you would agree that in the absence of enough evidence to put together a crude approximate distribution, it can hardly been viewed as a better alternative to just make shit up out of thin air ...
Pointing out the 77MM example is basically saying "Hey, look at this number! It might not be related to your shares or your life at all, but look at it anyway!" On the other hand, if you believe that 77MM is "reasonable" in some sense, well then you are absolutely talking about some approximate notion of a prior distribution on your exit value.
The salary distribution isn't even sourced. It was probably pulled from a source like salary.com which provides curves that look nearly identical to the one shown in the offer letter.
I don't think there are similar sources for more complex distributions (option payouts over all valuations).
It looks like the exit chart is designed to show what happens on the lower end of valuations (ie. as a lowly employee you don't participate). Also note that the X axis is serverely distorted: it ranges from $0-$100,000,000, but half the space is taken up by a blow up of a range of just $7,000,000!
Further still, the information on that page is not consistent. The chart refers to $2.35 options, but you are apparently granted $2.38 options.
The option payout calculator also doesn't appear to account for your costs to exercise your options either?
What's in the graph has only a nebulous connection to the example exit calculation. Those terms can change, and this still doesn't give any information on whether there is a preference multiplier or capped/uncapped participation.
There are also significant issues with retention / acquihire bonuses paid out of the acquisition price, severance agreements for employees who negotiate that they don't have to stick around, carveouts, and other forms of discretionary compensation -- none of which is addressed by either the chart or the page on the phantom 77MM exit number.
The point is that the current statement "only preferred shareholders are paid for exits below 25.7 MM" is not enough to actually capture the effect of preference and participation upon larger exits. If anything, it makes it even more confusing.
"One other thought: isn't it odd that you're willing to provide a distribution over salaries to argue for the salary you offered, but you just pull a number out of thin air for the possible exit?
Why not provide a distribution on exit values (and evidence for why you believe it), and then show the expected value along with risk, under that distribution."
I definitely see your point there, but that could also be seen as nudging dangerously close to "this is not an estimate of future returns... -wink, wink-". Better to be completely arbitrary in the number.
> Better to be completely arbitrary in the number.
What!?
No. If you want to go that route, it's better to simply omit any discussion about it. The worst thing is to present some arbitrary number that has priming and framing effects in the mind of the candidate. Especially if the number doesn't have a meaningful connection to the types of outcomes you might expect.
Providing a crude distribution or crude examples that imply a distribution is best. In the absence of that, providing no number at all is next best. Providing an arbitrary number is absolutely the worst.
Imagine if the chart showed an exit of 25 billion! That's also an arbitrary number, but it's clearly not legitimate if you can't back it up.
The fact that seeing 77MM doesn't hit you with the same "that's absurd" feeling that a 25 billion figure would should scare you ... because you don't have any information about whether the 77MM number is any more reasonable!
(Of course, you could just use base rates among start-ups, or start-ups in this industry ... but then you're using a distribution to justify things, so why not just actually use the distribution?)
You could go to the other extreme, if you're framing things (as they do) of "have no expectations" and give a valuation of $100.
That being said, I do see your point. It's hard to strike that balance between wanting to be more informative (and perhaps this could be separated from the offer letter) and setting expectations - it -is- important that people understand what their options may, or may not, be worth - but perhaps the offer letter isn't the place for that. Perhaps instead:
"If you require any clarification, this resource can explain things, or we can give vesting examples".
Transparency is a noble goal but a substantial part of this offer letter boils down to the counter-party in a financial transaction offering you financial and legal advice. That's a bad practice in transactions and makes this letter seem more like a sales tactic than something truly open.
The letter would be stronger if they just presented the facts (like revealing the cap structure, which unusual and interesting) and referred you to a neutral 3rd party for how to evaluate those facts.
By the way this letter seems to be incomplete because it has no reference to assignment of technical inventions or protection of trade secrets, which is de rigeur in most tech firms. There's also no mention of binding arbitration in the event of disputes. In the event the latter is not an omission and they do not attempt to impose arbitration I would say these folks are good eggs. The fact that they don't try to evade legal requirements would have more value to me than anything else in the letter when judging them as an potential employer.
75 comments
[ 2.4 ms ] story [ 160 ms ] thread2) The actual numbers there are a little bit scary. 0.03% equity grant? That's tiny. It's basically worthless.
When did startup equity get so tiny? I'm not singling eShares out here -- just broadly speaking. That means that if eShares exits past a billion dollars, those shares won't even buy a house in the area....
Non-technical line employee past Series A. See also the top 25% salary of $64.5k, which is roughly half of what new college grads get offered in engineering at AppAmaGooFaceSoft.
Fundamentally this is driven by one number which is an iron law in the Valley: the employee options pool gets 20% of the company. This is 2000 basis points. It will never, under any circumstances, suddenly have more than 2000 basis points in it. That's 2000 basis points you can issue to every employee from engineer #1 to social media marketing intern #6 in year 8.
What's the average equity allocation going to be when you have 400 employees? Well, it can't be 6 basis points, because 6 * 400 is greater than 2000. Is 400 employees a reasonable amount of a company at IPO? It's actually on the low side, particularly as companies are waiting longer and longer to IPO.
This means that, as companies get de-risked by succeeding investment rounds, equity available to employees drops sharply. Engineer #1 might get 200 basis points out of that pool, for giving up a sleepwalk-to-$300k offer from Google to take a flyer on a company paying $80k which needs to, ahem, build a CRUD app. (Not casting aspersions on any company by calling them CRUD apps. CRUD apps make the world go round.) That's still a low number relative to the amount of risk that engineer is taking and their opportunity cost.
I've heard on the grapevine that the iron law might be relaxed. Investors are pretty in favor of this, as long as the extra basis points come out of founder pockets rather than investor pockets.
The equity grant # was about #350 at a post Series B startup that would exit for about $1B later that year. My stake in the company was 1 basis point (0.01%).
Equity grants loosely follow an exponentially decaying pattern.
Having said that, you can only say there's no advantage in retrospect. If it had become a $10B or $100B company, 1bp would be substantial. And if the exit was a stock swap then perhaps the potential is still there.
2) I was a young desperate 20 something who didn't know better. ;)
You can find a (very small) list maintained by Zach Holman here: https://github.com/holman/extended-exercise-windows
I assume Zach got interested because he discovered after he was fired from github that he was close to losing all the money he "made" on options when github became a unicorn -- he probably had 90 days to exercise a very high priced illiquid security with horrid tax consequences if he fucked up. That's the thing about cash: it's not rigged. If you get cash, it stays in your bank account. Paper gains can disappear.
But as it stands right now, if you:
a) Are in SF or NYC
b) Have a CS degree from a well-reputed university.
c) Are working as a programmer.
100 k$/yr is not only achievable but practically the minimum.
That said, the linked offer isn't for an engineer, it's for a financial analyst, so perhaps 65 k$/yr is in fact market rate or better.
The only factor in that short list actually providing any indication of worth to the company is "has a CS degree from a well-reputed university" and that still doesn't indicate being able to perform in a job (rather than study).
I wish our economy could move beyond grading employees by degrees. They're proxies for the things that are hard to measure. Sadly I don't think they can easily be replaced (if at all).
Equity is a key component of modern compensation with better tax implications and more upside.[1] The “built in the garage” myth needs to die and doesn’t justify 100-1000x payouts when early employees are taking a significant risk too in opportunity cost. The main issue currently is information asymmetry as most early employees don’t see the cap table, don’t realize how much others own, don't understand how little risk most founders take, and don't know how much they are truly giving up in total comp from more established companies.
[1] https://medium.com/the-wtf-economy/what-paul-graham-is-missi...
Either way, I do not believe it to be a "key component" of "modern compensation". Tech companies can pay strong salaries just like other companies. They choose not to. That could change, and I think should.
Startup equity has always been tiny. Even for critical engineer hires 1-10. Engineers in general are poor negotiators and, in my experience, the employee pool is preserved for later executive hires that are brought in to bring the company and product into the fold of the industry trying to be disrupted.
Hell, just last week someone was on HN discussing walking away from $0.5m because he couldn't afford to exercise when he left an employer that was later sold.
If not, then the parent's point - a ninety nine per cent "sure bet" is still not a one hundred per cent guarantee.
What you really want, if you join at a point when you can afford to buy it, is to get options and buy and file 83b so you have zero tax owed.
This way you wont get the tax hit until the stock is actually liquid. The downside is that you cannot get longterm capital gains tax treatment, instead it's taxed as ordinary income with your income tax bracket.
Personally, I think all the tax optimizing on stock options doesn't seem worth it when most startup stock never see any returns. I wish IRS could find a way to fix this. It seems silly to pay taxes just because you own stock that is not liquid on any market and which you often can't even legally sell without company's permission.
In order to grant ISOs (Incentive Stock Options), which are taxed less unfavorably than NSOs (Non-qualified Stock Options), the strike price of the options must be equal to or above the "Fair Market Value" of the underlying stock.
The "Fair Market Value" is usually determined by an independent valuation firm through a process called a "409A Valuation." Firms like SVB, Alvarez & Marsal, Capshare and others do this.
The later-stage the company is, and the closer to IPO, the higher the 409A valuation will be.
If a company were to grant below the fair market value, there would be several bad consequences:
1 - You would owe taxes on vesting -- this means you could owe taxes possibly as soon as the first day of work, and likely on your 1 year anniversary of work. You owe these taxes whether you exercise or not;
2 - The company would have to withhold income taxes on your exercise date, including "employer-side" payroll taxes.
Normally, with an ISO exercise, you may face "AMT" (Alternative Minimum Tax) but you might not. By contrast, you certainly owe taxes with cheap NSOs.
Blame the IRS.
(This is not intended to be pedantic, but practical. You can vote on who sets the tax law. [at least most of us can])
Stock options are almost university priced based on a 409A valuation of companies "fair market value" which is a legal requirement. If anything various tricks are played to somewhat lower the valuation # spit out by these valuations.
Other than that, the letter is great and shows a wonderful commitment to transparency.
All the tech companies are talking about a shortage of skilled workers, but outside of management the terms of employment are still biased toward minimizing the company's potential costs. If there were a real shortage we'd see contractual changes.
To be fair this letter seems to be making an effort to be transparent, so I get what they are trying to do. As I noted in another comment though this is a tough trick to pull off successfully without seeming insincere.
However, is a normal distribution appropriate for income/wages? (Or better: Would this even have an effect on the percentile value?)
A policy to set salaries without considering how far above average the candidate is is unsettling. It speaks of seeing people as fungible cogs. If one person won't accept our distribution-based-standard-offer, oh well, next...
As a candidate I would see that and then begin to question whether other good people turned them down, and what caliber is the team I'd actually join.
Legally they can do up to 6 years.
All the tech companies I've seen use 1 year cliff/4 year total vesting. Ie, 25% after the first year, 1/36th of the remainder monthly.
But then I started playing devils advocate. If your sending an offer letter wouldn't you want it really personal and not as much about the $$$. Wouldn't you want to cite memorable and salient specifics about the individual.
This letter seemed like it was completely automated. It didn't seem "sticky" at all. I think a "personal" letter might have greater success particularly for smaller companies hiring expensive talent. Obviously you need to state what the candidate would be receiving for compensation but I think making it the only focus will lead to price comparisons.
However, I strongly support this concept overall. Often, I see the offer letter essentially presented as a formality once the basics are agreed upon. The candidate has completed the interview process, decided his/her non-financial interests in the opportunity, and received that "personal" positive reinforcement from the hiring manager & team. At this point: sure, the offer letter can include a personalized message to start, but ultimately it's for the candidate to review the hard facts: compensation, benefits, disclosures, etc.
I applaud this, but I'm cynical and think it would require an overhaul of the entire system regarding compensation transparency. The status quo offer letter simply benefits the company, and it's hard to justify changing course when only a tiny minority of companies are following suit. If this trend took off, then the threat of a competitive disadvantage come offer time would be a worthwhile motivator. I hope it happens someday.
Plus, there's nothing that stops them from adding a personal cover letter to this.
"We are excited by what you have to offer in the data science field - we are looking forward to seeing how you can ..."
that is really nice to see in an offer letter. a lot of hipster-y startups do the opposite, and act like its an alternative to hard cash.
The more I read this letter, the more it looks like they are trying to be selectively transparent in ways that don't actually cause them to give up any information advantage over the candidate -- and they are putting overly specific numbers in (like the magic 77MM figure for an exit) and tossing in relatively meaningless graphs, and an incredibly meaningless salary distribution, to add window dressing and make it sound like it's all much more technical, facts-based, and non-negotiable than it really is.
While it is a move in the direction of transparency, which is good in theory, this particular offer letter makes me feel like it's just a wolf in sheep's clothing -- still fundamentally trying to get a candidate to agree to be priced at a level well below their actual value to the company.
If the letter said "live your life as through your equity is worthless, because it is worthless and by agreeing to work for us for a lower wage you're not likely to recover those lost earnings down the road" .. then I might be impressed by the honesty.
I think their point was that the expected value of the company is approximately the same as the chance the company will Make It.
Obvious disclaimer: engineering at eShares.
And if shares are diluted or liquidity preference is added later after some fundraising, do you always go back and share all of the specifics of all of that with all employees? Kudos if you do, but if not it kinda renders the initial transparency less impressive or useful.
I've generally found the advice here [0] to be a much better proxy for hypothetical exit calculations.
One other thought: isn't it odd that you're willing to provide a distribution over salaries to argue for the salary you offered, but you just pull a number out of thin air for the possible exit?
Why not provide a distribution on exit values (and evidence for why you believe it), and then show the expected value along with risk, under that distribution.
At the very least, this way if your distribution over exit values isn't a severe power law drop off with uncomfortably low expected value, the candidate knows you're blowing smoke and can walk away.
Instead, you deny the premise that an expected value calculation is useful towards the end. Wow, that is absolutely walk-away-immediately stuff right there. "The distribution over outcomes is unknown..." Right, so of course you should make zero attempt to gather data about it and model it. That's one of the most bananas things I've ever seen, and it's right in the offer letter!
Picking a number like 77MM is almost like a dirty in-app purchase priming trick or something. (And no, it's not the least bit redeeming to add a tiny line at the bottom of the chart stating that it's just an example and not a forecast.)
If you personally, actually believe no one should engage in an expected value calculation regarding their start-up compensation package, then why not simply omit any slides about what the options will be worth? The 77MM thing is a red herring, and surely you would agree that in the absence of enough evidence to put together a crude approximate distribution, it can hardly been viewed as a better alternative to just make shit up out of thin air ...
Pointing out the 77MM example is basically saying "Hey, look at this number! It might not be related to your shares or your life at all, but look at it anyway!" On the other hand, if you believe that 77MM is "reasonable" in some sense, well then you are absolutely talking about some approximate notion of a prior distribution on your exit value.
[0] http://www.danshapiro.com/blog/2010/11/how-much-are-startup-...
I don't think there are similar sources for more complex distributions (option payouts over all valuations).
It looks like the exit chart is designed to show what happens on the lower end of valuations (ie. as a lowly employee you don't participate). Also note that the X axis is serverely distorted: it ranges from $0-$100,000,000, but half the space is taken up by a blow up of a range of just $7,000,000!
Further still, the information on that page is not consistent. The chart refers to $2.35 options, but you are apparently granted $2.38 options.
The option payout calculator also doesn't appear to account for your costs to exercise your options either?
There are also significant issues with retention / acquihire bonuses paid out of the acquisition price, severance agreements for employees who negotiate that they don't have to stick around, carveouts, and other forms of discretionary compensation -- none of which is addressed by either the chart or the page on the phantom 77MM exit number.
The point is that the current statement "only preferred shareholders are paid for exits below 25.7 MM" is not enough to actually capture the effect of preference and participation upon larger exits. If anything, it makes it even more confusing.
I definitely see your point there, but that could also be seen as nudging dangerously close to "this is not an estimate of future returns... -wink, wink-". Better to be completely arbitrary in the number.
What!?
No. If you want to go that route, it's better to simply omit any discussion about it. The worst thing is to present some arbitrary number that has priming and framing effects in the mind of the candidate. Especially if the number doesn't have a meaningful connection to the types of outcomes you might expect.
Providing a crude distribution or crude examples that imply a distribution is best. In the absence of that, providing no number at all is next best. Providing an arbitrary number is absolutely the worst.
Imagine if the chart showed an exit of 25 billion! That's also an arbitrary number, but it's clearly not legitimate if you can't back it up.
The fact that seeing 77MM doesn't hit you with the same "that's absurd" feeling that a 25 billion figure would should scare you ... because you don't have any information about whether the 77MM number is any more reasonable!
(Of course, you could just use base rates among start-ups, or start-ups in this industry ... but then you're using a distribution to justify things, so why not just actually use the distribution?)
That being said, I do see your point. It's hard to strike that balance between wanting to be more informative (and perhaps this could be separated from the offer letter) and setting expectations - it -is- important that people understand what their options may, or may not, be worth - but perhaps the offer letter isn't the place for that. Perhaps instead:
"If you require any clarification, this resource can explain things, or we can give vesting examples".
The letter would be stronger if they just presented the facts (like revealing the cap structure, which unusual and interesting) and referred you to a neutral 3rd party for how to evaluate those facts.
By the way this letter seems to be incomplete because it has no reference to assignment of technical inventions or protection of trade secrets, which is de rigeur in most tech firms. There's also no mention of binding arbitration in the event of disputes. In the event the latter is not an omission and they do not attempt to impose arbitration I would say these folks are good eggs. The fact that they don't try to evade legal requirements would have more value to me than anything else in the letter when judging them as an potential employer.