99 VC Problems But A Batch Ain’t One: Why Portfolio Size Matters For Returns
Practical Summary:
Most VC funds are far too concentrated in a small number (<20-40) of companies. The industry would be better served by doubling or tripling the average number of investments in a portfolio, particularly for early-stage investors where startup attrition is even greater.
If unicorns happen only 1-2% of the time, it logically follows that portfolio size should include a minimum of 50-100+ companies in order to have a reasonable shot at capturing these elusive and mythical creatures.
Like startups, most venture capital firms fail — at least in terms of returns.
Historically, 1/2 of all VC funds fail to return 1X initial capital. Another 1/4 fail to beat the (much more liquid) public market. Of the remaining top-quartile VCs that actually do perform, most can’t do it consistently across multiple funds. Yet we still view most VCs as pseudo-divine interpreters — powerful wizards who peer into their Palantir to see the future, tell us what new companies or trends will disrupt existing incumbents, and write big checks to amazing founders who create the next Insanely Great startup.
Except most of the time this is just a big bunch of baloney and they don’t.
For the few firms that by luck or skill get those predictions right, a strategy of very big bets on a very small number of companies can pay off handsomely. In fact, the more concentrated the portfolio, the better the returns will be for investors, assuming the portfolio still contains one or more big winners.
However, in its most extreme form, this strategy devolves into betting all one’s money on a single turn of the roulette wheel or buying a single ticket in a lottery. Surely winners of such games of chance should not be viewed as financial geniuses. Yet we still worship concentrated portfolio strategy as an industry best practice — when clearly, longitudinal performance of the venture capital asset class has yielded less than stellar results in the average case, and only consistent, frequent success for very few (~5-10%).
In a decade of investing in over 2,000 companies, I’ve found that a fewcompanies in any given portfolio perform extremely well, but they occur very infrequently. Most investments (likely 50-80%) don’t ever get to any kind of exit or return less than 1X invested capital. Perhaps 15-25% of portfolio companies succeed and result in a small exit of 2-5X. Another 5-10% might attain valuations of over $100M (which we call “centaurs”) and achieve exits of 10-20X. And if you’re lucky, 1-2% attain valuations of over $1B ( “unicorns”) and result in very large returns of 50X or more invested capital.
In summary, most investments fail, a few work out OK, and a very tiny few succeed beyond your wildest dreams.
While these numbers might be unique to my own experience and process of investing, most people in the industry would not disagree that large outcomes happen infrequently, or that a few big investments tend to dominate returns (re: Peter Thiel / power law, etc.). If this is true, then a more prudent VC investment strategy would be to construct portfolio size based on number of investments required to generate at least one big outcome (or ~3-5 large outcomes, to be on the statistical safe side).
Currently, most larger VC funds ($200M+) doing Series A/B investments rarely invest in over 30-40 companies, and most micro-funds (<$100M) doing Seed and Series A investments rarely invest in over 50-75 companies.
The current VC fund industry average portfolio construction is inherently and critically flawed – undersized by a factor of 2-5X.
I believe a more rational number of investments is ~50-100 companies for later-stage funds, and at least ~100-200 companies for early-stage funds.
Let’s presume that even for the average khaki-wearing VC — tall, smart, good-looking, went to all the right schools, and likely very white and male — that their portfolio distribution looks something like this.
Simple VC Fund Model: Unicorns @ 2% and Centaurs @ 5%
Looks pretty doable, right? With only 7% big wins, a VC fund could theoretically return almost 2.5X — hey, we should all become VCs!
But given the frequency of centaurs and unicorns (5% and 2%, respectively, in this model), let’s look at three portfolio sizes of 15, 30, and 100 companies.
What becomes excruciatingly clear from this simple model is that returns are dramatically based on the number and % of unicorns (and possibly also centaurs) that occur in a portfolio.
If portfolio size is too small, you risk finding ZERO big wins.
www.correlationvc.com
If unicorns occur only 2% of the time (or less), then with portfolio size of less than 50, you might not find any. In fact, given the drama that occurs with many startups and VC funds, you might come to the conclusion that you really don’t feel statistically “safe” without constructing a portfolio that gives you a decent shot at 3-5 unicorns. So unless you think you’re going to pick unicorns at a rate of 5-10 % instead of the 1-2% industry norm, you are essentially GAMBLING with LP money by selecting a portfolio size of less than 50-100 companies.
Depending on how often the average VC is able to find and invest in unicorns, a minimum safe number for most large funds is 50-100 companies, and for early-stage seed funds (where company attrition rates are even higher), a “right-sized” portfolio requires at least 100-200 startups. Possibly more.
This is a very simple example — management fees and other expenses are not included, nor have we modeled any follow-on capital typically reserved for 2nd/3rd check investments in winners, nor have we modeled the timing of deployment and return of capital, nor any recycling / re-investing of capital.
However, the basic argument remains. Portfolios too highly concentrated in a small number of companies risk missing out on ANY unicorns whatsoever, under-performing and perhaps even failing to return 1X initial capital.
Dave has been a nerdy Silicon Valley entrepreneur and investor for over 25 years. Prior to PVC, he was founding partner of 500 Startups, and previously worked at Founders Fund and PayPal. Dave has invested in hundreds of companies around the world, and some of them aren’t dead (yet).
Dave has been a Silicon Valley entrepreneur and investor for over 25 years. He has invested in hundreds of startups around the world, including 10+ IPOs and 40+ unicorns (Credit Karma, Twilio, SendGrid, Lyft, The RealReal, Talkdesk, Grab, Intercom, Canva, Udemy, Lucid, GitLab, Reddit, Stripe, Bukalapak).
Prior to launching PVC in 2019, he was the founding partner of 500 Startups, a global VC firm with $1B AUM that has invested in over 2,500 companies and 5,000 founders across 75 countries. Dave created 20 VC funds under the 500 brand and invested in 20 other VC funds around the world.
Dave began his investing career at Founders Fund where he made seed-stage investments in 40 companies, resulting in 4 unicorns and 3 IPOs. He led the Credit Karma seed round in 2009 (acq INTU, >400X return). His $3M portfolio returned >$200M (~65X) in under 10 years.
Before he became an investor, Dave was Director of Marketing at PayPal from 2001-2004. He was also the founder/CEO of Aslan Computing, acquired by Servinet in 1998. Dave graduated from the Johns Hopkins University (BS, Engineering / Applied Mathematics).
Nerd Cred
Forbes Midas List, top 100 global VCs (2016, 2017)
Largest multiple on invested capital: Talkdesk @ >1200X
Largest realized return on invested capital: Credit Karma @ >400X
Street Cred
Dean’s List (2X), Johns Hopkins University
Academic Probation (2X), Johns Hopkins University
Slept through final exam, Circuits & Systems (oops)