the next trillion dollar company will be a partnership

ethos: the purpose of writing this was to document why the way textql will be run will become the way that all good companies will need to operate. a more opinionated version of this detailing why i emotionally prefer this and why someone should work with us will be written here - but this is more analytical. feedback is requested on axioms that feel… weak.

skeleton

1. the best engineers are getting more leverage

2. because of capital liquidity and general productivity trends

3. higher leverage ICs need partnership-like exposure to be retained and build big things

4. these will look like partnerships for other industries with high leverage ICs

5. system partners are going to destroy traditional tech firms.

the best technology ICs are becoming 2 orders of magnitude higher leverage & proportionally harder to keep

the cost to attract and retain the best science and engineering talent in the world has been exponentially going up over the past 10,000 years. it once required not imprisoning them at home or setting them on fire for heresy.

in 1957, the stakes were raised by fairchild semi, attracting the brightest by promising them no suits and intellectual freedom. netflix raised the stake with top of market pay. google raised the stakes with professional chefs and ping pong. facebook invented RSUs to give them more equity in the venture they were working on. ramp raised the bar still - by enabling 23 year old college drop outs to command a staff engineering title and $700K compensation…

and still - ramp’s finest are leaving left and right to start their own companies. if you were one of the first 50 at ramp, retool, rippling, scale, palantir, or stripe, you’re basically always evaluating your current job against the opportunity to swing at a $7b venture of your own. not everyone is leaving in pursuit of monetary upside, but intellectual achievement and clout are intricately tied with the financial upside as a founder. today’s best companies have no way to compete against that opportunity cost.

well, they have one way. scale, ramp & even openai have compensated by acqui-hiring their most entrepreneurial talents back into the fold after a year of working on their own thing. in the process. the brief year they spend on their own venture rewards them with usually a P&L of their own, a promotion, and an outsized equity grant that those would-be-founders would’ve never been able to negotiate for in the traditional career ladder. they secure an “acquired founder” title.

but this deterministic script is stupid. a year of productivity is gone. thousands of legal fees are set on fire to fund, facilitate, and acquire the small startup. there was no new information gained by the would-be-founders… other than “year 0 is hard.” there’s no new information gained by the acquirers - they knew how valuable the talent was before they left the door. there’s no information gained by the VCs - wow a great founder who recommended these great ICs to you ended up wanting them back. big woop.

this whole process is a meme to get around the fact that companies today don’t have a mechanic to give founder-level exposure to engineers who command founder-level leverage.

this is happening because capital is rushing into venture and general output has gone up & concentrated

so what changed? two things:

1. there’s way more capital availableeveryone knows this already, but the scale is worth noting. when fairchild was first started, there were ~4 VC funds in operation. by 2000 there were hundreds. today, there are thousands.more money means more companies. more companies means more founders. if there are 100x more companies being funded - that means it’s 100x more common for someone to be good enough to be a venture backed founder.which means the bar goes down as more capital seeks out talent earlier and earlier on [mckinsey presenting in high schools & jane street sponsoring IOI competitions is an example of the same phenomenon]to these VCs - a 25 year old at ramp being paid $700K a year is a potential unicorn that can return their fund. increasingly, ramp now has to compete with the fact that their engineers can raise $2M on 20M on a whim. suddenly, $700K doesn’t seem that good anymoreaside: the intelligent reader here might notice that strong tech operators are a scarce resource… and if there are now 100,000 companies instead of 100 competing for them, it’s unlikely that this scarce resource just 100x’s… and instead there are probably fewer strong operators per company… which might explain the drought of trillion dollar companies founded after 2000… but we’re not there yet.

2. the productive output of the 100 engineers compared to the median has gone way up, and people recognize that engineering output peaks at around 4 YoE [doesn’t go down, but asymptotically plateaus]as far back as 1975, people have known that the top 10 engineers in a set of 1000 can complete a project in 1/2 the speed of 1000. two orders of magnitude more humans only yields 2x more speed. this is the best engineering talent is asymmetrically productive [insert 10x engineer meme]. additionally, dependencies [and the need to communicate] scale exponentially with the more humans you have on a project, i.e.

1. this is compounded by the growing gap between the top 1% and median engineer. more capital in tech means more jobs to fill… and more jobs to fill means bootcamps are churning out 3 month sub-par talent to write sub-par-non-scalable code that they don’t communicate for, bc increasingly software engineering is being seen as a “meal ticket” for folks. this brings down the floor & medianthe scarcity of strong engineering talent [pulled away to found, across more companies] multiplied by the effect of work-from-home and work-life-balance culture w/ low un-employment means managers are increasingly doing bullshit work and have no ability to hold a high output bar for their team. [wall street firms like Salomon Brothers exhibited a similar effect in the 1980s. wall street firms like Salomon Brothers no longer exists.]ai then serves as a force amplifier still. people on twitter have been glorifying the ability for anyone to spin up a project from scratch from ai-generated code, but as everyone knows: it would take a new grad an afternoon to build a functional twitter clone. the hard part is in the complexity of scaling it. lines of code are things you spend, not things you produce…or so i’m told, disclaimer: i don’t know shit about engineering, i have 7 MBAs and possess no skills to speak ofbut my peers tell me that giving github copilot to a sub-par engineer is like giving 10x leverage to a wallstreet-bets trader. ai is an amplifier to strong engineers, but is a tech-debt-multiplier to weak ones.the most striking example of software engineer leverage going up is companies like ramp and cognition are credibly competing with high frequency trading firms like jane street and HRT for talent. for a startup with 10 employees to be competing on financial upside with a firm famous for paying new grads $550K annualized on their internship reflects what the most concentrated teams value each engineer at.

the structure of technology firms will shift to a model that uses exposure to retain high leverage ICs

ok. everyone’s pointed out that the number of google / facebook / microsoft tier companies has gone down over time. the biggest IPOs of the last bull market were airbnb, snowflake, uber, and all of those companies are still trading at around ~<$100B in market cap

what happened?

i have no empirical grounds for believing this, but it seems like every time i check w/ one of my friends who used to work at retool or scale, they say the best people there when they were around have all left to start companies. the next generation of break-out companies seem to have a shorter and shorter time horizon on how long they can keep their best ICs engaged.

https://twitter.com/CamdenKo/status/1757101090804547741

if you were andy jassy at Amazon in 2006 in today’s funding environment, what incentive would you have to create AWS, making Jeff Bezos 1000x more wealth than you, when you have so little exposure… if you can just get any VC to write you 20 at 200 post to start aws yourself? btw jassy’s not even a billionaire

empirically - it seems obvious that the previous generation of trillion dollar companies were built on the backs of contributors who saw their upside extracted by the founders and investors in those companies. that’s why the andy jassy’s of this generation are just leaving to start their own thing. anecdotally, every single one of my peers often talk about how fucked the first 10 employees at any startup get relative to the upside / risk ratio of the founder. none more so than the ex-founders who i’ve gaslit into joining our team.

https://twitter.com/varunramg/status/1772340796622049577

but starting a trillion dollar company is hard - takes decades at least, and along the way you can’t bleed your best people. and if they leave - they’re going to run into the same problem you did.

given all of this - the next trillion dollar company will be built in a way that gives your best team members access to the thing they care the most about: which is founder-level exposure to upside. this doesn’t mean everyone makes millions - the same way raising from a VC doesn’t guarantee you millions - just the opportunity to get that asymmetric upside on a risk-adjusted basis.

we can speculate about the structure of that, but it’ll likely look like a general manager type model where they run their own P&L and command something like equity, future royalties, revenues, or profit.

these firms will look like other industries w/ high leverage ICs, like banking, PE, law, consulting

there are other industries where individuals regularly command very high levels of leverage: private equity, investment banking, law firms, or consulting firms. these institutions are often structured like partnerships, but since they’re designed around the needs of those individuals, structures can vary creatively. some examples:

investment banking firms

investment bankers have high variability year-to-year, as their revenue is the result of fees charged for deals executed. macro has a role to play, but every single partner demonstrates their value year to year by selling their rolodex services, with a power law often kicking in. rainmakers like andre mayer and felix rohatyn at lazard were pulling down 80% of the revenue between them - and commanding 10% of the firm’s revenue as compensation at its peak. these rainmakers get an outsized voice in determining compensation distribution for partners.

if you earn, you earn the right to determine compensation for others

mckinsey

at firms like mckinsey or pwc, advisory partners manage a more consistent book of ~$20M in revenue each year. partners are added horizontally without vertical leadership - and are in charge of managing the associates and analyst below them. partners must by in, and following the example of james mckinsey - sell their shares back to mckinsey at book value when they retire. the managing director of mckinsey is elected by the partners once every few years - and voluntarily steps down when they’re not the most well suited for the role.

michael milken at drexel

exposure to profits is why michael milken never considered starting his own venture outside of Drexel. despite operating the most powerful arm of drexel, he didn’t feel a need to spin up his own house because he could divvy up the profits of his junk bond division as he saw fit. - he had enough exposure and was able to capture ~$6B w/o ever needing to be a founder

the origin story of blackrock

in 1988, larry fink didn’t want to risk his own personal capital to start asset management firm. steve schwarzman [founder of blackstone] didn’t want to give away blackstone equity, so in exchange for a $5M line of credit, blackstone obtained a 50% stake in blackstone financial management. today, that institution is known as blackrock, and manages 10 trillion in assets.

while software engineering is traditionally thought of as “high leverage”, the revenue commanded per employee at google or facebook is only around ~$1.5M. this pales in comparison to the $20M every mckinsey partner is expected to bring into the firm. today, the best engineers are JUST starting to rival the leverage commanded by a general partner at KKR or cravath partner.

of course, there are going to be challenges in structuring R&D as an partner w/ revenue exposure - when it’s not directly revenue generating. there will need to be more modular ways of valuing engineering output. it’s not obvious why software assets couldn’t be treated like other assets like real estate, just make the asset amortization faster and treat it like super-high maintenance costs.

these partnerships could be more decentralized or more centralized, and the unit of incentive can vary greatly. engineers could be compensated with royalties on future revenue of a piece of software, future profit, exposure to the P&L, a liquid internal equity-like market that benchmarks the value of a subsidiary companies stock against the parents.

these partnerships will be more agile and efficient, dominating the incumbent tech-giant model… the best ones already look like this

higher talent density. lower communication load. direct connection to the output of their work. partnerships formed by these engineers will be able to retain the best people for decades and harness the efficiency of partnership models.

illustrated through example

imagine being the GM of AWS S3. right now, your path to promotion lies through hiring a ton more people so you can empire build and declare you deserve to be VP. your amazon stock is nice, - but your performance one year is drowned out by some dumbfuck hiring thousands of indian workers to mechanical turk no-touch checkout at go stores. you being more apathetic about the cash efficiency of s3 seeps through the team - resulting in apathy across the org.

now imagine you’re directly paid bonuses on the profit of s3. every single person who reports to you is another salary eating into your bonus pool. you feel your comp directly tied to every piece of efficiency. you stop counting team size and take pride in revenue per employee - retaining only the most elite of your team to rewards them like kings, and fighting aggressively for the best talent on the market.

it’s already happening.

the most efficient companies in tech are already structured like this

• rippling hires ex-founders to be general managers who own a horizontal slice, and have direct competitors outside of rippling to power the compound startup model

• scale has general managers that run P&Ls on verticals, scaling up and down their budgets based on the amount they’ve grown in the past quarter, with the ceo allocating budgets like a VC investing rounds

• palantir expects every customer to grow to $10m/yr, causing them to deploy forward deployed engineers as “CTOs of the specific customer” - and draw resources as demand grows

• netflix gives wide deference to team leads to set their own policies and hires only the higher leverage engineers w/ top of market pay. the 10x leveraged exercised by the best is explicitly referred to in their culture docs

better talent. more upside. more exposure. work harder.

they’re going to run circles around incumbents.

there will be a company that produces 100 billionaires.

…

i’ll do everything in my power to give you the power to earn more money than any other company will pay you. text me hi at 5105618393