Do VCs Overrate 'Ex-Unicorn' Founders?
Do venture capitalists overvalue name-brand startup experience? I was recently at a “Future Founder” dinner hosted by a VC where I considered this question given nearly everyone invited works or worked at some well known startup.
Having a name-brand experience (e.g., Ex-Stripe/Ex-Uber/Ex-Ramp) certainly helps you in getting hired or in fundraising (Full disclosure: I have a few on my resume!). Similarly, it’s become more and more fashionable in VC to fund entrepreneurs who don’t even have an idea yet purely based on their credentials, which often includes experiences at these types of startups (example program).
But at risk of tarnishing my personal brand (whoops), I’ve come to believe these experiences are not as valuable as VCs think. First, the data shows that founders of the most successful technology companies did not tend to have this type of background. And second, my personal experience working at one of these name brands (Uber) showed me they might not be the best at teaching the right lessons about building a successful company.
The best tech founders ever don’t follow this path
To investigate this question, I looked at the founders of the current top 10 US tech companies by market cap and excluded companies started before 1990 to make it most relevant to today (see appendix for more).
If I had to categorize what these founders backgrounds were, I would put them in the following buckets:
- Extremely smart students who had no prior work experience
- Mark Zuckerberg (Facebook), Larry Page/Sergey Brin (Google)
- Former founders whose previous startup was less successful
- Elon Musk (Zip2/PayPal -> Tesla), Travis Kalanick (Red Swoosh -> Uber), Reed Hastings (Pure Software -> Netflix)
- Smart people who had a great idea but didn’t work in Tech previously
- Brian Chesky/Joe Gebbia (Airbnb), Jeff Bezos (Amazon)
- Tech veterans (mostly at big companies) with subject matter expertise
- Jensen Huang (NVIDIA), Marc Benioff (Salesforce), Fred Luddy (ServiceNow)
Notably, while there are many founders of previous tech companies, there's barely anyone with the background of an “early joiner” of a high growth tech startup from which they used their knowledge and experience to build their next thing (maybe Benioff but it’s debatable).
This general theme matches my personal experience. The most impressive founders I’ve personally worked with had no experience in tech before they started their companies.
What about the PayPal Mafia?
Are there exceptions? Of course. A very notable one is “PayPal Mafia”, which was a high-growth startup that had several early employees who later went on to found even more successful startups. Many of these startups don’t show up in our initial list either because of acquisitions which took them private (ex: LinkedIn acquired by Microsoft, Youtube acquired by Google) or they were successful but not enough to be in the list above (Palantir, Affirm, Yelp, Reddit).
But for all the mentions of PayPal Mafia, the example seems somewhat unique. It had <50 people when acquired by eBay, much smaller than most hot startups people leverage today when fundraising, and leaders at PayPal seemed to have a uniquely strong eye for talent. And in the 25 years since then, many other proclaimed “Mafias' have failed to produce similar results. The alumni of more recent hot startups (Uber, Stripe, Tesla, Airbnb, etc.) have produced many startups with a lot of funding but I have yet to see similar successes if we look at public companies.
It’s still early for a lot of these “Mafias”. And I am also open to the argument that the best founders of today’s generation will have different backgrounds than startups founded 20 years ago, which show up at the top of the most valuable public companies today simply due to having more time to grow. But I am starting to think the PayPal Mafia says more about the ability of certain people to find talent (Peter Thiel, CEO of PayPal) than something inherent to the benefit of working at high growth, successful startups.
Experience at name-brand startups often doesn’t prepare you for being a founder
My own experiences working at a name brand startup (Uber) and meeting folks in tech with similar experiences have also reinforced my view. It’s not clear to me working at these companies actually teaches you much about how to start a successful startup. In some ways, they may even teach you the wrong lessons.
I believe finding product market fit (PMF) is by far the hardest part of creating a successful startup. But the vast majority of employees at startups join post PMF, like I did when I joined Uber in 2018. Joining Uber later on teaches you a lot about “growth-stage” problems like how to best grow while improving profitability, or what’s the right regional strategy to pursue. But it didn’t really tell me much about how to navigate the “idea maze” in the first place and find PMF, as that was found very early on in Uber’s life.
In some ways, an experience like Uber can actively mislead you about how to build a company because it makes things seem easier than they are. Uber had arguably the greatest product market fit of all time. There were problems to solve, but at some point Ridesharing was simply inevitable.
Many of the problems I worked on were competition problems (ex: Uber vs Lyft). This was because it was not a question of whether this product/business would work (given the very strong PMF), but instead who would win it. Most startups die though not because they lost to competition but because they never had product market fit in the first place.
When startups become status, name brands attract other name brands
Another reason name brand startups can be misleading as founder factories is that they tend to attract the wrong kind of people to be successful founders. In the last decade, working in tech/startups has become a high status thing to do. MBAs in the 2000s wouldn’t think of joining a tech startup, let alone founding one. In recent classes, significant shares do both.
Once name brand startups become name brands (i.e. post PMF), they then attract the other name brands to them (BCG, HBS, etc.). These types of people are often extremely smart and capable, but I don’t think they often fit the founder mold we identified above that has seen the most success historically. They tend to be more risk-averse and have higher opportunity cost, which I’d argue makes them less likely to stick with startups that aren’t going well.
It still may be a good decision for many of these people to have a go at entrepreneurship and there are plenty of successful examples of course. But on average, I’d guess this group of founders will lead to more disappointing outcomes than most investors think.
How to find the best founders?
So how do you find the best founders then? I’m no VC (and never have been), so I’m obligated to speak with some humility. But my main point would be that in today’s world, ideas have become underrated.
Ideas have fallen out of favor in the last decade. It has become more and more common nowadays for incubators or VCs to fund smart first-time founders who haven’t even chosen an idea, trusting they’ll figure it out with time. There also isn’t as much focus on investing in founders who have experience in the domain of the company they are starting, which was the one example from our list above where former tech employees were successful (Benioff, Huang, and Luddy all started companies in the domain they built experience in)
This doesn’t mean background doesn’t matter or that it’s not a plus to see certain experiences on a resume. Successful founders need to check a lot of boxes, and there are plenty of cases of smart people repeatedly pivoting until they find their great idea. But I’ve seen too many people who are very smart and hardworking spend a lot of time on ideas that just didn’t work.
Appendix
Top 10 US tech companies by market cap today (founded after 1990). I exclude companies which are spinoffs (ex: Broadcom) whose parent company was founded before 1990. Companies pulled from here, all metadata (ex: founding date) pulled from a GPT extension to Google Sheets.
When there are multiple founders, I choose the most relevant one (ex: Zuckerberg, Benioff) but the results hold up even when including cofounders.
I pulled the same list of top public companies founded after 2010 to see if that changes things. From a quick look through, there are more potential examples (ex: Brian Armstrong of Coinbase worked at Airbnb for one year), but I believe the results still largely hold. There are even more examples here too of people with deep subject matter expertise starting companies in areas they know (Snowflake, Zoom, Crowdstrike, Robinhood, etc.)