In 2011, Benchmark Capital led a $12 million Series A into Uber. Three years later, they put money into Lyft. Two competing companies, same lead investor, same fund. Uber’s founder Travis Kalanick reportedly went ballistic. The relationship between Benchmark and Uber was already strained by that point, but the Lyft investment made it personal in a way that startup lore tends to remember as drama and forget as strategy.
It was pure strategy.
Understanding why requires letting go of the idea that venture capital is in the business of picking winners. It isn’t, at least not in the way founders imagine. Venture capital is in the business of owning the outcome of a market, and those are very different jobs.
The Setup: One Market, Multiple Bets
Benchmark’s position in 2014 was not unique. Sequoia has backed competing payments companies. Andreessen Horowitz has portfolio companies that compete directly in enterprise software. This happens constantly, and it is not a bug in the system. It is the system.
Here is the financial structure that makes this rational. A typical venture fund returns most of its gains from a small number of investments. The industry rule of thumb is that a handful of companies in a fund of twenty or thirty generate the bulk of the returns. The rest return little or nothing. Given that distribution, the correct behavior for a fund is not to pick one horse in a race. It is to own the category.
If ride-sharing is going to be a hundred-billion-dollar market (which it was), the question from a portfolio perspective is not “will Uber win or will Lyft win.” The question is “do we have exposure to this market at all, and how much.” Backing both Uber and Lyft at different stages, with different ownership stakes, is a rational hedge against the uncertainty of execution. Companies that look like sure things in 2011 have a remarkable tendency to collapse under their own weight by 2016.
What Actually Happened
The Benchmark-Uber-Lyft triangle played out over years and got genuinely ugly. Benchmark eventually sued Kalanick in 2017 to force him off the board after a series of governance disasters. They alleged he had concealed material information during a stock authorization vote. Kalanick called it a coup. Benchmark called it necessary. The suit was eventually settled.
Set aside the personal drama for a moment. Look at what Benchmark actually held at the end of that period. They had a massive stake in Uber, which went public in 2019 at a valuation that, despite a rocky IPO, eventually made their early position worth billions. They also held a position in Lyft, which IPO’d two months before Uber. Two exits from the same category. Two shots at the same macro bet paying off.
From the outside this looks like disloyalty. From inside the fund’s return calculations, it looks like competent portfolio construction.
Why Founders Get This Wrong
Founders tend to read a term sheet as a vote of confidence. An investor picks you, commits capital, joins your board, and becomes a partner. The relationship feels personal because it is personal, in the early stages especially. Investors spend time with you. They give advice. They make introductions. The whole performance of venture capital is designed around the relationship.
But the relationship exists inside a financial structure that has different incentives. A partner at a fund has a fiduciary obligation to their limited partners, the institutions and endowments and family offices that put money into the fund. That obligation runs upstream to the LPs, not laterally to the founders. When a VC backs your competitor, they are not betraying you. They are doing their job, which was never what you thought it was.
This is not cynicism. It is just an accurate description of the incentive architecture. The failure to understand this leads to exactly the kind of governance disasters that produce the Kalanick scenario: a founder who treated investors as allies until the moment the investors behaved like investors.
The Market Ownership Logic
There is a specific calculation that VCs make that founders almost never see. It goes roughly like this: if I believe this category is real and will generate large outcomes, I want to be in it. If I can only be in it through one company, my return is dependent on that company executing perfectly. If I can be in it through two or three companies, I am essentially long the category rather than long a specific team.
Being long a category is lower variance. You give up some upside on the winner (since you own a smaller percentage of it), but you dramatically reduce the risk of being completely wrong on execution. Given that execution is the hardest thing to predict about an early-stage company, this trade is usually worth it.
The catch is that this logic works better for markets where winner-take-all dynamics are less severe. Ride-sharing turned out to have significant winner-take-most effects in individual cities, but the US market ended up supporting two viable national players. A VC backing two competing social networks in the early 2000s would have found that the dynamics were harsher.
Sequoia’s position on this is worth noting. Their fund documents have historically included language that explicitly permits investment in competing companies, and they have defended the practice on the grounds that their job is market exposure, not exclusive partnership. Whether you agree with that framing depends on whether you are a founder or a fund manager.
What Founders Can Actually Do With This
The practical implication is not that you should distrust your investors. It is that you should be clear-eyed about what the relationship is.
When you are taking money from a firm, ask directly whether they have portfolio companies in adjacent spaces and whether their policy permits investing in direct competitors. Most will answer honestly. Some will give you a vague answer that tells you something. Get it on record if you can. Board seats matter here: an investor on your board with a stake in a competitor has an obvious conflict of interest that should be addressed in governance documents before it becomes a crisis.
More importantly, understand that the best protection against getting managed out of your own company is not the loyalty of your investors. It is building something that clearly needs you to run it. Kalanick’s situation became terminal not because Benchmark backed Lyft but because Benchmark had a legitimate governance argument when things went sideways. The leverage existed because the company had real problems that the board could point to.
Venture capital is a financial product with a warm face. The warm face is real. The financial product is also real. Founders who understand both sides of that equation are harder to surprise.
Benchmark’s investment in Lyft was not a betrayal of Uber. It was a read on how large transportation markets resolve, placed before anyone knew for certain how they would. As it turned out, they were right on both counts. That is the business.
