Which business schools are more likely to produce unicorns and unicorn founders?
What did unicorn founders study at university?
To create a unicorn, what is the optimal number of co-founders?
To find answers to these and other questions about companies that achieve elusive billion-dollar status, go to … LinkedIn?
That’s right. The most rigorous, reliable data on unicorns and their founders is likely found on the LinkedIn profile of Ilya Strebulaev, a finance professor at Stanford’s Graduate School of Business who posts weekly insights from his dataset in sharp, graphically pleasing detail.
DATABASE CONTAINS 531 UNICORNS
“Venture capital research is relatively recent because there really hasn’t been any data. It hasn’t received as much attention as I think it should have received. I have amazing colleagues at other schools — at Chicago Booth, at Harvard Business School, at MIT – who are doing research on venture capital, but it’s a relatively small group of people,” says Strebulaev, a 2020 P&Q Professor of the Week.
In 2015, Strebulaev founded the Venture Capital Initiative at Stanford GSB to pull together researchers and compile more reliable data. He and his team of research assistants, PhD students, project managers, and lawyers have poured through tens of thousands of pages of VC contracts and other documents, data sets from PitchBook and VentureSource, LinkedIn and other websites, news articles and more in an effort to compile a database that includes every U.S. unicorn startup since 1995. That’s 531 unicorns so far.
“Most of those research posts on LinkedIn that you see — and many, many more are coming — are really with the support of the VCI,” Strebulaev says.
HUNDREDS OF LIKES & SHARES ON PROF’S UNICORN POSTS
After sharing insights from the dataset to his MBA students in his popular venture capital class, he figured others might be interested as well. He shared the first unicorn LinkedIn post in October: a cartogram showing the weighted-average location of American unicorns by state. (See above. Not surprising that California is the big orange blob with 309 unicorns headquartered in the state.)
A couple times per week, Strebulaev uses the dataset to answer pressing unicorn questions:
- What’s the right age to become a unicorn founder? (30-34 years)
- Do unicorn founders need an academic degree? (It helps)
- What is the founder gender of VC-backed startups between 1991 and 2018? (94.5% male! Come on, ladies! Let’s turn those ideas into companies, already.)
Strebulaev’s posts generate hundreds to thousands of likes, shares and comments. Poets&Quants talked with him recently about unicorns, unicorn founders, and the high interest in his novel data set. Keep scrolling for answers to more unicorn questions along with an Instagram announcement for P&Q readers. (This interview has been edited for length and clarity.)
First, tell us a little about yourself.
I was born and educated in Moscow, but I moved to London and did my PhD in finance at the London Business School. I came to Stanford as a professor of finance in 2004. I’ve now taught more than 2,000 students, mostly MBAs.
About 10 years ago, I became very interested in venture capital – not just about startups, but really about the entire innovative ecosystem. This is the ninth time I’ve taught the venture capital class which is now becoming very popular. Not just because of me and my co-teacher – Brian Jacobs, co-founder of Emergence Capital – but because the topic is very, very popular these days. Obviously, I’ve been impacted by my students and, in fact, most of my interesting research questions start with a question that a student asks that I don’t know how to answer. Some of them lead to really interesting discussions, debates with other colleagues, and some of them lead to fantastic research papers.
I read that collecting this data often means pouring through reams of contracts and other documents. It seems like a very labor intensive process.
It is extremely laborious. We actually have a team of lawyers who help us to interpret particularly challenging contractual details. In the Anglo Saxon legal system, contractual partners are allowed to innovate within the broad legal framework and precedence. As a result, we have tens of thousands of venture capital contracts, and really every single contract is unique. We devised a system where we have about 100 variables that we’re collecting data for from these contracts, and we have instructions that are 100-plus pages long on collecting them. Some of the variables are easier, like quantitative variables for the number of shares. Others are much more complicated, and have to be reconciled.
One of the reasons that venture capital research hasn’t gotten as much attention is that all those companies are private. Some are very small, and many of them are going to fail inevitably. So, many of these companies just disappear from the planet. To collect data from something that doesn’t exist anymore is very difficult. Many of the companies don’t really have a lot of financial data, and even if they do, there’s no one centralized place where you can look for it.
I think the VCI is in a nice position where we have gotten access to some of the data – not all the data, I would like to have more of the data – that is representative of what’s happening in the universe. Contractual data is interesting because we have a large representative sample of contractual arrangements across the venture capital industry.
A lot of the LinkedIn posts I’ve seen so far have looked at unicorn startup companies. How are you defining unicorns for the purpose of the research?
The definition of unicorns that we use is those companies that had at least one round with a reported valuation of $1 billion. When I say ‘reported’ it is important because it is not a fair market value. It is for a private company’s post-money valuation.
There are two types of unicorns: One is a company that has raised $1 billion or higher post-money valuation in private rounds. Another is a company that exited a private status – either it went public or was acquired – and, in doing so, had a valuation of over $1 billion. That is my definition. There are various other definitions used, and many people use the word ‘unicorn’ without having a precise definition.
NEXT PAGE: Academic degrees earned by unicorn founders + Business schools that produce the most unicorns
Why are unicorns an interesting topic of research?
Well, first, there is a correlation between unicorns and success. If you look at the most successful venture capital backed outcomes over the past 20 years, almost all of them are unicorns. I think in the venture capital world, over the past 10 years, a unicorn is considered like a badge of honor, especially if you reached unicorn status as a private company.
I think many also think that ‘unicorn’ is identical to the word ‘success’, which is not the case. We know unicorns that failed. For example, in our data set, there is Theranos which, as you might know, is in the middle of a jury trial right now. My data tells me it was a unicorn — at some point, during its history, people perceived it as a very successful company.
What I think is also interesting is not just looking at unicorns, but comparing them with all the other companies that will not become unicorns. On my LinkedIn posts, I thought I would start with the unicorn sample because, frankly, I thought they would generate more attention. But, we’re also looking at what I call a random sample or representative sample of similar companies, but not all of them reached unicorn status. So this would allow us to compare those companies that became, at some point, highly rated with those companies that never became highly rated. And those posts will be coming to LinkedIn too.
In one post, you looked at the academic degrees achieved by 1,263 founders of 521 U.S. unicorns. Of those less than 5% were college dropouts, but only 18.7% earned MBAs. (Compared to 38.4% of bachelor degrees, 20.5% of masters and 22.6% of doctorates.) Is the MBA still valuable for startup founders? (See graph above)
Well, I’m biased, so I do think that an MBA is a valuable degree for all sorts of reasons, not just because you can become a unicorn founder. But that is one of them, of course. Trust me that many of my students at Stanford are obviously taking my venture capital class because they would love to found a unicorn. And, by the way, some of them did.
If you have an MBA degree, and you founded a unicorn, or if you’re a student and your classmate co-founded a unicorn, apart from being pretty exciting, I think it’s very important for networking reasons. We do know, not from my research but from many of my colleagues elsewhere, that the business school network is extremely important.
In some other posts, you looked at which business schools had produced the most unicorn startups and the most unicorn founders. Harvard came out on top with 44 unicorns and 53 unicorn founders.
Yes, we looked at the number of unicorns and unicorn founders for various business schools, which doesn’t take into account size. That shows Harvard Business School in first place, Stanford GSB in second, and Wharton School in third. I can actually give you the full list of schools for all the business schools for American unicorns. (See table at the bottom of this page.)
But, when measured by unicorns per capita, Stanford comes out on top.
If you look at just the number of unicorns at business schools, then Harvard Business School comes out on top because Harvard is much larger than Stanford GSB. Harvard Has 53 unicorn founders and 44 unicorns in my dataset while Stanford has 33 founders and 23 unicorns.
But if you look at the number of founders who earned a business graduate degree on a per-capita basis, Stanford GSB comes out on top. Stanford has 3 unicorn founders per 1,000 MBAs while Harvard has 1.6 and Berkely Haas has 1.4.
What about Stanford, do you think, produces more unicorns (on a per capita basis?)
Well, first, Stanford is the birthplace of Silicon Valley. As they say in real estate, ‘Location! Location! Location!’ Of course, Stanford did not locate itself to be in Silicon Valley, Stanford participated in the creation of Silicon Valley.
Second, I think what is really important is the relationship between Stanford business school and its alumni. We have extremely strong alumni come to campus and hire graduates. We have an amazing startup garage, for example, where a lot of alumni come and help. We also have people who have become very successful and they come back to teach, they spend a lot of effort imparting their wisdom. What I really like about my classes at Stanford is the combination of what I hope is rigorous academic knowledge with real practical inputs from the best practitioners. I think that’s something that we do well at Stanford.
The third thing is, the network is critically important. When you come to the Stanford Graduate Business School, even people who’ve never had any entrepreneurial experience of any kind think, ‘Well, I should do something,’ because everybody around them is doing something.
NEXT PAGE: Age of unicorn founders + What comes next
Who's the audience for this research?
That's a very good question, and it is why I started the LinkedIn posts. As an academic, my original, most important audience is my fellow academics – people who know the methodology, know the data very well, the economic reasons and so on.
But when I started the Venture Capital initiative, I wanted other people to become aware of this as well. The first audience I thought about were policymakers. I think policymakers in this country are really unaware about the importance played by startups, innovation, and venture capital in long-term economic growth.
Then, I’ve had a chance to meet many aspiring entrepreneurs who don't have access to Stanford MBA classes, and I realized that this knowledge would be interesting to them. I think some of the value is that it is partly inspirational. I mean, if you look at the comments people make on my posts, that's what they say. Maybe academics care less about that, but if you can show that you don't need to fit a very specific profile to be successful, that there are many profiles for success, that is inspirational.
Now, because I'm talking to Poets&Quants, another big audience I have in mind is students and aspiring students of business schools elsewhere. If you're at business school and reading Poets&Quants right now, the very first thing you should do is to go to my LinkedIn profile and have a look at the posts. (Laughing)
My goal is to post at least twice a week. I have enough posts to last many, many months, and I hope that people find them useful, informative, but also inspirational.
Does this research live anywhere in particular? Is there a website people can go to, or just follow your LinkedIn?
We are now in the process of forming something, but right now there are many LinkedIn posts on the profile. There's also some posts on LinkedIn Articles, which aren't as popular these days, but are a much more detailed description about how I got a certain post.
A lot of people reached out to me and suggested a newsletter because a newsletter would allow us to go into much more detail about how the data was sliced, caveats to the data and so forth. I haven't done that so far, but I'm considering it. Again, this is the start of the journey, and if readers have any suggestions about how to make this more informative for them, I would love the feedback. Feel free to email me.
The LinkedIn posts are really about the research, I think, but I just recently -- and you're really one of the first people I'm telling about this -- started an Instagram account where I'm posting slides from my MBA venture capital class.
What’s coming down the pike? What questions will you be asking and posting about in the coming weeks?
An interesting series we’re going to post soon is more about unicorn founders. We're going to post about academic majors of unicorn founders, which I think is really interesting. We're going to post about the ethnicity of unicorn founders, and then at some point later, we're going to post about immigration status, which I think is also extremely interesting.
Next we will be posting about unicorn outcomes. So what happened to all those unicorns in our dataset? I'm interested in coming back to unicorn founders and looking at what they did prior to their unicorns? Did they work at venture capital backed companies? Did they work at large companies? At small companies. Then finally, I will have several more posts about corporate venture capital. So my LinkedIn plate is quite full.
Watching your LinkedIn for the different ways you parse out the data has been interesting and pretty fun, actually. But why do you think venture capital research is important? Why does it matter?
Well, I think this matters because the largest companies right now in the United States were created with venture capital. So, I think startups -- not just unicorns -- along with financing and funding innovation are really critical components of our future. If you look at the 10 largest publicly traded companies in the U.S., seven of them didn't exist 40-50 years ago. All of them were venture capital backed. My prediction is that 20-30 years down the road, we will have the same number -- 7 out of 10 -- of the largest companies created through venture capital backing. This is, of course, pure speculation. Some of them may be companies that right now are very small startups. Some of them may already be unicorns, and some of them maybe have not yet been born. Some of them may be born this year or next year by one of those people who hopefully read my LinkedIn posts.
These companies have a huge impact. I should think that the impacts of venture capital and Angel financing is much much larger than has been appreciated by many -- including by the policymakers and by society in general. I think the future is going to be determined in a large way, by the way this industry is going to operate. Well, that's my conviction. I think that is really the main reason, at the end of the day, why it's important.
The post Which Schools Produce The Most Unicorn Founders? This Stanford Prof Has Crunched The Numbers appeared first on Poets&Quants.