> Throw out the old cap tables. A founder doesn’t get 30% and an early engineer shouldn’t get 0.25%. Those are old numbers from when you had to raise VC capital before you could build a product. Before everyone could and did start a company.
Choice quote. I'm kind of amazed this isn't brought up more.
Agreed. This shift hasn't taken root yet and I think you're going to see a lot of startups that succeed caught on to the fact that the extra early-stage equity they can retain (due to less onerous funding) is best given to the first N employees than held onto by the founders. Taking the equity the VC's normally would have gotten and giving it to your first several employees now opens the door for you to hire people who are at the top of their field as your first employees, if they are not totally risk averse. I think it might turn out we end up inventing a new term for these "post founder, pre employee" folks since they are going to be a big differentiators for this generation of startups.
There are definitely stages in the life of a startup that are pre-salary but less risky than ground zero. For example, some initial market validation, users, or even paying customers could be there. Or, there could have been a significant engineering breakthrough or two. Or, simply the fact that the startup has survived its first pivot is evidence the team has staying power.
especially if you've had an ethical bypass and are willing to go Zynga on them later. Engineers are a lot easier to push around than angels (or VC's) are.
completely agree. as someone that's just about to graduate, i have to say i was a little shocked when i figured out this is the way things are. the difference between the first engineer and one of the founders never seemed so categorical to me, and this was one of the few things that made me feel it's not really worth joining a startup as early as I'd like (to work on interesting+useful things, have impact, etc.) unless I'm actually involved at the founding stage.
Yes this is one of the main reasons why I am not working for a startup currently even after running my own startup for 2 years and having loved the experience. Over the course of interviewing with multiple startups and getting offers the amount of equity being offered was so little that the companies would have to hit ~250 Million valuations for the options to be worth in the mid 5 digits. From my perpective the risk-reward ratio for the first 10 engineers is much much worse than that for the founders. I do intend to go back to founding something again in a couple of years.
I'm not saying your general intentions are wrong, but you should be able to get a much better deal from startups than you have seen to date if you want. I'm moderately surprised at your experience.
I cant name and shame the companies but I was surprised about the offers. EDIT: fwiw - for the purposes of the original post I am not including offers that I got where I was offered a higher stake but where I did not have a high conviction about the startup.
That might be true but I find 0.2%, thus $500k for a $250M valuation, still very low.
$250M valuations are quite an achievement already and $500k, while it will pay off a mortgage, is not life-changing. (not talking out of experience here :)) So, an early engineer who will most likely be very instrumental to bringing the company to such a valuation ends up with a(n admittedly very) nice bonus, while founders do end up with life-changing wealth.
Ya, I was just saying that is sounds like dman wasn't even getting "standard" offers. I wasn't really weighing in on whether the "standard" offer makes sense or not. It's a challenging question.
Typical to what? What is offered? I'm not sure why we're even still talking about what is offered since it should be clear by now that early employees won't be getting that anyway.
> the difference between the first engineer and one of the founders never seemed so categorical to me
The categorial difference is that if the company flops in a few months the founders are out a lot of money, whereas as first engineer you are merely in a similar position to before you took the job, but with some nice experience on your resume and a few months worth of pay in your bank account.
Most founders I know would pay themselves a salary after the first funding round. A small one perhaps, but I don't think the difference is as big as you make it out to be. Especially when you consider that the first engineer in a startup gets at least the same workload as the founders.
That assumes there is a round of funding. Not all startups are in the kind of sexy areas that cause VCs to throw money at them. Some are funded by the savings of the founder (and sometimes a second or third mortgage on his house). At that kind of startup, the founder often only gets paid his salary if the company has profits.
>whereas as first engineer you are merely in a similar position to before you took the job
Not at all. An Engineer can only have one job at a time so they most likely quit one to get this one. Now they have nothing, no income at all. Some founders will be in a similar position but many still won't have to get an office job if their startup fails.
Of course, less the quite possibly considerable difference between market rate and the reduced rate the engineer is being paid. If you go by Fred Wilson's numbers [1]: $10k per month per engineer, the engineers are probably accepting a $60K/year salary. That's an enormous pay cut for me at least, and well below market for virtually all engineers. So certainly the founders are taking very little money, but the engineer may well be making a $4k/mo investment as well...
$10k/month x 12 months/year. Isn't that $120k/year? How did you come up with $60k? This is a serious question, because I assume there's an unspoken assumption in your post.
That's the fully loaded cost of the employee, so the sum of salary, payroll taxes, health insurance, unemployment insurance, office rent, and whatever else I'm forgetting.
I'm curious what Paul Graham and other top VCs think about this. This would be a pretty big shift in how companies expand early on, and it could potentially change the personality of the first employee.
After the demonstrations that money not in-hand is likely to evaporate before you can get it, is that still the case? Personally I ignore equity completely when deciding if a position is well paid or not. Equity is like a lottery ticket a friend buys me: I'll keep it around and when the time comes I'll see if I won, but I'm sure it's actually worthless.
I suspect that just means you don't want to help someone start a business where it is hard to measure how effort in generates money out. If a buddy shows you a picture of pirates burying treasure in your backyard and then hands you a shovel, it isn't a big stretch to imagine you digging a hole for a few hours for no pay :). On the other hand, an idea scribbled onto a napkin by a non-expert isn't so compelling without cash to go with it.
Your comment hints at the issue. It isn't measuring effort in vs money out, it's likelihood of reward for effort. At this stage it's clear that if you count options and even equity as zero you'll often be right.
If a startup hands me a picture of treasure in some place I can quickly get to (i.e. little investment on my part) and a shovel, I'll invest a couple of hours for the chance of getting rich. If they hand me a lottery ticket that they can void if it did happen to win, I'm not going to spend months or years slaving away for that.
> There isn’t a shortage of developers and designers. There’s a surplus of founders.
This is especially true of non-tech companies looking to hire quality people for positions in their IT departments.
I currently work in such an IT department and we have a position for a J2EE developer to write portlets for our up-and-coming Liferay portal. Needless to say, we can't find anyone good (outside of recruiters, which we'd rather not use) to fill this position. My boss has asked each of us if we know anyone we can recommend for this position. As far as I know, no such luck.
Personally, I'm not the least surprised. None of the developers that I know would even touch a job like this. Most don't/won't do J2EE. Portals never really took off like corporate America hoped it would (and Gartner said it would). I have some things that I'm working on on the side, and if/when that becomes something, I'll be headed out the door too.
What my employer fails to understand is that they are hiring as if it was still 1998. Developers now have options and no longer have to settle for jobs like this. Their competition is no longer other IT departments, it's freedom of choice itself.
So what's the answer when jobs like this need to get done?
It may not have to be J2EE, but there are plenty of corporate jobs that have to get done to help build big company infrastructure, and if much of the talent that was available now has more attractive options, is the only answer to escalate salaries? Or to hire in shorter-term contract developers?
Well, if good engineers don't want anything to do with your opening at the compensation you're offering... increase the compensation.
And, honestly, change the position so that the person hired isn't considered just a cog in management's plans, who's expected to be on call 24/7 and to put in 50 hours a week in a windowless room. I wouldn't go back to that shit even at double my current salary.
Agreed. The reason people put up with soul crushing jobs like investment banking is because the compensation makes up for it. Granted, most people in those kind of jobs burn out after a few years, but you anticipate that when taking the job. I have several friends doing the "make tons of cash while young, move to easier job in 5 years" career choice.
Corporate America IT jobs have all the soul crushing without the reward.
It's not really burn out, it's the management cabal that likes to keep the body count high. There's a lot of mildly desperate people who really forced their way up into unstable career positions, and natural genocidal tendencies peek out every now and then. It's innate human behavior. History's purges were not by accident.
By portals, I mean enterprise portals (like Liferay, JBoss Portal, Apache Jetspeed, etc.).
The biggest problem with them is that the "applications" or "portlets" that occupy its pages we're really meant for dashboard-like information. Just show the most relevant information in a small space on the page. If you want more detail and functionality, you should be able to click on something and be taken to a full-blown website.
In reality, IT managers here the word "applications", and they then try to cram the full web app into the portlet. So now you're stuck trying to provide web app functionality in an environment where the only programmatic links you can generate are to yourself, you have no ability to control what other things are going to be on the same page as your app. Forget being able to set HTTP caching headers, since you aren't in control of the page.
A lot of these Java-based portals have plenty of open standards. A lot of JSR-### compliance.
From my perspective as a developer, I don't use the document sharing or publishing features of Liferay. My focus is on the portlet development. This means JSR-168 and JSR-286 portlet APIs.
I know nothing about SharePoint, so I can't really compare the two. I just know that I really dislike Java portlet development, and am ready to never have to do it again.
I feel the urge to point out that, if the cap table gets squeezed, there is no a priori reason that any particular part of it should get squeezed. It is entirely possible that, as he frequently argues, the productivity of money is going down (because "startups are cheaper than ever to start") and the productivity of product teams is up (because of huge amounts of leverage in the system via OSS, platform companies, improved development technologies, The Cloud, etc).
If one buys that set of facts, there is exactly one participant in the startup ecosystem who should be getting told "Sorry, your contributions are not worth what you think they are." It isn't founders or engineers.
P.S. That said, psst psst, being last cofounder beats first engineer 100% of the time.
The productivity of money may be going down but I don't think it has much to do with the prices of startups.
The main thing is that there are relatively few opportunities outside startups to make a good return on investment -- the western world is still (again) in a recession, stocks are so and so and what you can get on US fed bonds is a joke.
So really there is too much money chasing the opportunities available.
So the good times will continue until the economy gets back in good shape.
Which I for one don't believe will happen for at least a decade.
One assumption being made here is that hiring is a critical task. It's entirely possible that a company can achieve its objectives without hiring; if that's the case then the argument doesn't apply.
I actually don't understand the author's point or numbers. A fully loaded engineer costs 100-200k depending on where you're located in US/Canada. And let's say you need 25-50k for 2 founders to live for a year. Giving yourself a 12 month runway you're at needing 150-300k in the bank to even be in the position to hire.
If you have 150k in the bank it's because you have some seed money, or have bootstrapped your way to that position. If you've bootstrapped then you've achieved pretty good product/market fit and have good revenues. Conversely if you've raised 150k and haven't achieved product/market fit, then you probably shouldn't be hiring.
Now, if you look at companies that have raised 150-300K I don't think you'll find many of them hiring, at least not very aggressively. I feel that level of money is used to give the founders time to iterate until product/market fit and not be cash constrained when it comes to things like contracting a good designer or buying a domain. And once a seed funded company achieves some semblance of product/market fit they go and raise 1-2 million.
So is this article intended for the few companies that have raised in the 100-300K range that are starting to hire? If that's the case then ya, I agree such a company is going to have difficulty hiring. They have significant risk of not achieving product/market risk, and they don't have sufficient capital to give a fair market wage. So a potential employee is taking a huge chance on such a company, and should obviously be compensated for that (relative to the very attractive compensation packages are companies like Yammer, Square, etc). But I don't really think there are very many companies like that, but maybe I'm out of the loop.
Before: Founders put together a biz plan and raised a decent seed round and then started hiring the first few employees.
Now: Some person gets an idea, pitches it to another person, and they start building it, then they pitch it to a third person. They don't pay themselves salaries, live off their savings and their credit cards. An 'incubator' may give them some cash to pay for things like AWS instances and filing incorporation papers. They may be up to 5 or 6 people before they have a 'minimum viable product' (MVP) and are willing to pitch it to VCs for a real series A.
The author points out that all of the first 3, 4, 5, or 6 people who were working to get the company to the MVP point, they are founders. Not just the 'idea guy' or the 'wizard' that they snagged to help implement the idea. Everyone who came on board before series A has made it possible to get to that point is a founder.
And yet there are companies that treat person 2 - n as 'employees' and give them way less equity. Because of that people don't want to be employee 2-n, and thus hiring them is 'hard'.
I've noticed this as well, and have been puzzling around with the following thoughts.
Lets say you create a company and decide that prior to series A, 80% of the company will be owned by the founders and 20% to outside investors. You start with 10 shares, 2 for an angel, 8 for the founder.
Now you add a founder, you double the share pool and now you give 8 to the new founder and 2 to the angel (distribution is 4 + 8 + 8). Now you add another engineer/founder and you now add 80 shares to the pot and distribute them 27 + 27 + 26 for the founders and 20 for the angel. Add a new angel and you double the shares to 200 where each angel get (20 + 20) and founders get (54 + 53 + 53) shares.
The idea being to keep the ownership percentage of the company 20% angels, 80% workers.
Now you go for series A - my thought is you pick three ratios, investors/founders/employees. That may end up being 49/31/30. Allow your angel to either contribute their shares to the series A (cash out) or to participate. But at the end of the day the representation is 49,31,30. Same deal when you add more investors, add to the size of their pool so that works, add to the other pools to keep it balanced.
I am undecided if it would make managing the equity table easier or harder.
Just to nitpick a bit -- your final three numbers add to 110, not 100. Perhaps you meant 49/31/20? I point it out only because 49/31/30 seemed too good to be true.
This rings very true. I've interviewed with and turned down 8 YCombinator start-ups in the past three years. Of the 8, I would have been within the first 5 employees of 6, within the first 10 of the remaining two. They all offered below market salary in exchange for "huge equity". Thoe "huge equity" offers ranged from 0.35% to 2% (the 2% one offered me 40% of market-rate). I didn't mind, I was only interviewing to network and try to sell my $150/hour consulting services to them. (3 of them are still customers). Being an employee for a start-up is a sure-fire path to an early grave in potter's field.
This is why I don't consider jobs with even exciting startups with early traction at this point in time. Why would I bother getting 50x less return for a similar investment and risk as a founder? We should at least be in the same order of magnitude. Especially if the startup is practicing "lean" and is going to completely change by the time it exits. Might as well just start my own company.
And, if you can't get into an incubator, you can just as easily bootstrap your way through the early stages.
I see a lot of positive comments for this post but I'd like to chime in and suggest that upping the equity is not going to solve the problem.
In once sense the engineer is making the same kind of bet as the VC. They don't know the founders too well, not sure if the idea is any good, its a big gamble. Buy unlike a VC, the company will be the engineers whole life until the idea proves itself or fails.
A few extra percentage points of a deal that will probably never happen is not really all that attractive.
The only solution is to raise more money and pay them more.
"Raising the first $25K for product development is easy – join an incubator."
I think that trivializes the process. There are now multiple incubators, but about two orders of magnitude more looking for spots. Y Combinator gets thousands of applications for a few dozen spots. Tech Stars Boulder got 600 apps IIRC, and only 10 get in the program. Early stage money gets trivially easy if you're in one (especially YC or TS), but getting further funding is still a time drain, and now you're competing with the 100+ incubator-based startups for VC/Angel money.
I do agree about cap tables. With less money needed to start a company, the equity normally given to investors can (and should) go to employees. It's an opportunity for many who have talent to get a bigger reward for their hard work.
I suspect those numbers may be skewed by the same "we only hire the top 1%" phenomenon. There could very well be a recurring group of people with not so great ideas that repeatedly apply to incubators, and repeatedly get turned down, whether with the same idea or a new one.
I'd be interested to see the skew as well. Maybe a large part of those declined were unserious college students or similar.
Sure, at least half of the applications hit the trash bin. I recall a comment from pg at startup school in 2010 correctly, he said a surprising number of the applications to YC (I want to say 20%) were in ALL CAPS. That's an easy way to self-select one's way into the bit bucket. But even at half the size, there's still at least 1000 applications that are likely well-formed in one way or another. PG has said that the hard part is finding something exceptional in an application that would put it over the top.
I couldn't raise 10K to build a prototype. Had to earn the cash the ol' fashioned way. Unless a handshake and a smile can get money from someone you know, raising cash is frikkin' hard.
Oh, and offer me as much salary as you can with the option for me to dial that back into equity. I can't pay a mortgage or buy pizza with options, and my market rate does not fluctuate based on how much buzz you have on twitter.
The founders' got Pauli as an engineer, any problems, he goes to Pauli, problems with EC2, he can go to pauli, troubles with rails, hard drives, Linux, he can go to Pauli, but now he's gotta come up with Pauli's money every week, no matter what.
Your tweet didn't go viral?
Sorry to hear about that, fuck you pay me.
Your funding didn't come through?
Fuck you, pay me.
People don't need more coupons?
Fuck you, pay me.
Servers crashed in the middle of the night?
Fuck you, pay me.
And then finally when there's nothing left and he can't raise another dime from the VC, or clear anymore paper on second market, you bust the joint out, take the code and release it as open source.
If founders actually believed their option bullshit it would be a no brainer to keep those valuable options and pay engineers 200K. There is no shortage of engineers, only a shortage of suckers.
Your goodfellas analogy is pretty close to perfect (and I know quite a few paulis myself, hell, I'm a Pauli). That being said, the author's suggestion of a vastly larger amount of equity with the right founders in place might convince me. But .25? Even 3? Fuck that, pay me.
I think VCs usually value successful startups much more highly than engineers do, so I'd probably raise more from VCs (which has some other benefits), and pay higher cash compensation (or other compensation, like profit sharing/revenue sharing, benefits, nicer office, etc.) in place of above-market equity grants. It really depends on the individuals, though.
When there is a scarcity of a product (engineers to hire), the cost naturally goes up. If you are having trouble hiring, raise your compensation. If you don't have cash to offer you need to raise the amount of equity you give (or perhaps other perks).
Sometimes I feel like the last person left who raised traditional angel money and then VC money.
We paid our first engineer market rate and had actual benefits. Our cap tables are pretty traditional and we haven't had much trouble hiring all things considered (though we are extremely picky).
I have to say, if you have to give 20% of the company to your first engineer in order to get him/her on board, then I don't really see the benefit in going the incubator route over traditional funding.
The issue with this method is that it assumes that the investor model has to change too. VC's will attempt to have the right ownership over the lifetime of an investment eventually getting to around 20%. Some want 20% right away. The problem with granting so much equity to employees makes the rounds of funding required later difficult to match this ownership target.
Investors want founders with meaningful ownership of the business, a syndicate that owns enough to care, and employees happy too. That is where the model is driven from, and would require a change at the investor level for this to work.
Assuming 2 investors (40%) 2 founders (40%) that leaves only 20% remaining which does not fit the proposed changes.
I really appreciate the comment about the surplus of founders (as opposed to a shortage of developers.) There's a context that's very important in that statement: that not all founders are really necessary, nor do they really come before other key team members (namely, engineering) in terms of foundership.
This article helped me clarify an interesting contradiction in my own motives. As a founder I'd much rather pay market rates and hold on to equity - 0.5% incentive doesn't hurt me but my hackles raise at 10% or more. (The state of my hackles is not correlated to how fair the compensation is, BTW - it's just greed.)
On the other hand, even more than getting the top 2% unicorn squad, I want engineers who believe in the business. Taking equity in lieu of salary is a sort of screen for this, and I think that's why I've structured things that way in the past.
I agree with the article that it's not fair to the first N employees, though.
My question is, if not the traditional equity model, what would you offer to get and retain key early engineers? How about sales people? Community manager?
To clarify, perhaps people can post their thoughts on what equity percentages you'd offer to the following? (or something similar) :
I think that this is true in Silicon Valley and possibly NYC, but not in other areas.
I firmly believe we just haven't churned out enough CS graduates nationwide over the past decade to meet the demand of both the startups and the established businesses that are all hiring right now.
But the mobile social coupon CRUD owners only want to hire the top 2% and then cry they can't find anyone. The job market is like taking a racing horse and having it plow a field. The incubators have a similar problem, the candidates are so good they don't even really need the incubator, I imagine the incubators doing worse as time goes on because people are not so hungry for success and will just combine with others when the going gets tough.
Or, just maybe, many of the startups synergizing twitterscapes and being the X-but-social-networked might not need high-caliber bleeding-edge CS grads, and could probably get by with coffee-fueled hacks?
You don't have to be a CS graduate to be good at programming.
You need to have a genuine interest in what you're doing. Otherwise, you're not likely going to do (let alone enjoy doing) what it takes to be good at this.
You always need to improve your current abilities and add new ones. That can be done by coding and reading (probably in that order). Github is great for this kind of thing.
You need to be aware of advancements and changes in your profession. Sites like HN are great for this.
I agree that something in the cap table has to give, but I don't think it's founders. It's investors.
Psychologically, I would also make sure those "early employee founders" really risk something; even if it's just securing $20K of their savings in a bank account to be used in a pre-approved way if the company needs to make payroll. $20K is, after all, just 2 months of gross salary for your self-determined great engineer applicant.
Doing something like this has 2 benefits:
1. Applicants self-select for risk tolerance.
2. You avoid the spoiled kid syndrom of "co-founders" making all types of employee requests ("I need $6K for a top notch working station", etcetera.)
Thank you for showing and communicating a sense of clarity that many colleagues around me (sometimes myself incl) seem to lack. People talk of 'founding' a startup and 'i'm an entrepreneur' as if its like a roadtrip to tahoe. Its not. Period.
Instead of comparing the equity of the first engineer with the founders, it is easier to compare how much his lower salary can buy him if he takes another job with full salary and invest the difference in the company as an angel investor. If there is an engineer whose salary is $120k/year and is joining a startup at $90K/year, he is taking a 30k/year loss. Let's say the startup has received $500K investment at the valuation of $2.5M. Since the startup has passed its valuation point already, its current value is somewhere between 2.5M and its future expected value.
Let's assume the prediction for valuation at the next round is $10M and there is 50% chance that the company gets there. This would make the value of the company about $5M at this point. This means that the engineer's discounted salary is worth about 0.6% equity (30K/5M=0.6%) for the company. Usually companies make the offer for 4 years worth of equity with a vesting plan. Again it is not correct to multiply 0.6 by 4 because the salary of engineer will reach to its market value after the next round of funding. It is fair to multiply it by 2x. This brings the total equity given to the engineer to be about 1.2%. I made a few assumptions here such as what the expected value of the startup would be in the next round of funding and how much the salary is lower than the market value. This calculation shows that the current amount of equity offered to the first employees is not that different from what it should be contrary to what the author of the blog post has suggested.
To offer a simple formula:
Y (expected equity of the first employee for the first year) = X (loss in income for the first year) / V (valuation at the next round of funding) * P (probability of the startup getting to the next round).
For our example:
Y = 30K / 10M * 0.5 = 0.6%
I know engineers often compare themselves to the founders and wonder why they should get so much less equity considering that they have similar skills and are putting equal effort into the company. One thing they ignore is that what founders have already put in. In most typical startups, the founders have been developing the idea at least for two years and have worked full-time on the startup for 6 to 12 months before receiving the seed funding. They have done this at the time that the possibility of getting to the seed funding round was less than 20%.
If we assume their market value was $120K/year, that means they each put in something about $120K at the time that there was less than 20% chance that the company would get to the point of $2.5M valuation. If there are two founders, this would be about $240K investment at the valuation of $500K (2.5M * 20%). That means the founders should get 48% in vested shares in the company. Instead they get all their share as unvested shares and have to work for the next 4 years in the company to earn them. Considering the remaining sacrifice they have to make, it is totally fair for them to receive 60% instead of 48%.
Don't underestimate the hard, fractal, handwavey problems, though. Place a thousand apes together and they start forming chaotic systems that can be very hard to understand. What's worse, most bugs on that front are virtually guaranteed to be undiagnosable without altering the dynamic—and there's ethical problems with just getting out the logic probe…
Choice quote. I'm kind of amazed this isn't brought up more.