Pattern Matching
The Venture Capitalist’s Dilemma
Photo by Andrew Ridley
Few jobs are as coveted as those in venture capital. Promises of riches, prestige, and fame abound. Yet these enticements belie the realities of the job. Indeed, the work is hard, success is evasive, and the demands couldn’t be higher: generate outsized returns without the benefit of real-time (or near-real-time) feedback loops. To the extent that “feedback” is accessible, it may be outdated. To perform their duties amid a challenging, evolving environment, investors turn to pattern-matching: the focal topic of this piece.
Pattern-matching, broadly speaking, is an observation and decisioning methodology. It deals with matters of perceived correlation, treating yesterday’s stories as a prelude to tomorrow’s truths. Among investors, its use is widespread, with some calling it their “secret sauce”. Yet its adoption isn’t limited to that audience, for even founders, aware of its utility as a “signaling” mechanism, use it to source talent, attention, and capital.
The influence of pattern-matching shouldn’t be overstated. Its power to steer capital and confer legitimacy upon a chosen few is unrivaled. That it can reliably typecast the persona of entrepreneurial success only highlights its growing influence over who gets to be legible in the marketplace of ideas. Predictably, pattern-matching triggers strong reactions. Its detractors argue that it’s too exclusionary and an unreliable predictor of entrepreneurial aptitude. Its proponents counter by citing its successful track record, claiming that it necessarily de-risks the investing process. What’s lost in this dialogue is not the zeal with which each side argues. It’s that both perspectives, when taken together, don’t naturally conflict. Indeed, both can be true.
Yes, investing in startups is hard. And within reason, of course, appropriate steps should be taken to de-risk the investment process to stimulate returns. And yet, the operative word is “appropriate”, and when that threshold is unmet, returns are likely to suffer. Further, the erasure of promising entrepreneurial talent is yet another concern that merits attention. At scale, this malpractice only serves to jam the faucet of innovation, creating an intellectual dam of sorts — one that stymies the flow of fresh ideas, products, and discoveries. This, in my determination, is the troubling duality of pattern-matching in its current form. What was designed to strengthen the investing process has worked in some respects but fallen short in others. This gives rise to the Venture Capitalist’s Dilemma: Investing in precedent-defying founders and teams, which data suggest are good for performance and returns, deviates from Silicon Valley investing practices — thereby leaving scores of talent on the sidelines.
How pattern-matching is done needs to be challenged — for if not to protect the idea that talent is evenly apportioned, then at least to protect the performance of the asset class.
Photo by jesse orrico
Humanity + Science = Pattern Matching
Humans are innately social. Bonding is at the core of what we do, though we may self-select into isolated, exclusive spaces. Our proximity, or lack thereof, to others shapes our networks, behaviors, and what we do and don’t see. And as we age, it’s harder to forge new bonds, bonds that would otherwise free us from our near-homogenous enclaves.
In addition, humans tend to embrace conformity. This is neither good nor bad, of course — for to some extent, our ancestors found utility in this practice. Nevertheless, when conformity extends too far, it inspires groupthink, a condition that chills dissent and penalizes deviations from the status quo. The latter runs counter to the ethos of Silicon Valley. It’s also the anti-recipe to achieving power-law-driven returns — the outgrowth of placing correct, contrarian bets on the future — which is arguably the singular mandate of risk capitalism.
Furthermore, humans are creatures of habit. We adopt routines, swapping uncertainty for consistency, variability for regularity. Our longing for predictability stems from our longing for control. To attain it, we observe. We search for clues — the vestigial footprints of truth. In turn, we analyze, measure, and experiment. We want our realities to bend to our desires, less so to those of others — and especially not to chance.
Finally, pattern-matching is premised on prescience, which is driven by the concept of science: the domain of observation, experimentation, and prediction. Silicon Valley is forged from science, and pattern matching borrows from that tradition. Indeed, science proceeds by testing the merits of a hypothesis; undeniably, the same is true of venture investing — or at least it should be. Investors use investments to evaluate theses, and future returns, if any, serve as a proxy for their success. That these two fields share much in common is not a coincidence. Rather, pattern-matching is born out of the scientific process, and both fields seek to unearth “truth” — albeit for different reasons.
Yet practicing empiricism — or finding inspiration from it — warrants studying representative data sets: the inputs that undergird the truth-discovery process. Failing to do so births the creation and perhaps the uncritical adoption of flawed, unrealistic models that unfortunately shine under the compromised banner of “legitimacy”. Machine learning experiences this challenge, and an argument can be made that the venture capital industry does as well. Pattern-matching, when based on closed, insular networks only produces much of the same: insular, exclusionary patterns that lack completeness, and fail to depict the breadth of the entrepreneurial community. At scale, this threatens innovation, progress, and the performance of the asset class.
“if A then B” ≠ “if not A then not B”
Historically, the go-to “pattern” in venture is young, white, male, computer science, Stanford, MIT, and Berkeley, etc. To be fair, that playbook can work. There are scores of brilliant technologists and entrepreneurs that emerge with and reflect those profiles, as evidenced by the following chart:
We should celebrate their wins, for they are our own. And we should encourage more of them to throw their hats into the entrepreneurial ring. But the presence of those successful profiles should not be taken as evidence that only individuals with those profiles can find outsized success. Such an assertion is fiction at its purest, necessitating a challenge not only to that dubious assumption but also to those who willingly or unwittingly accept it as gospel worth spreading. Brilliance is not a function of gender, ethnicity, or Alma mater. It is also not a function of geography. Genius is ubiquitous. There is not a single identity that will or has ever cornered the intellectual or creative market. As such, it is imperative that pattern-matching changes to fully recognize the intellectual and entrepreneurial talents that surround us.
Readjusting the Pattern
Here are some thoughts on tweaking pattern-matching for the better. The aim is to reduce the likelihood of missing amazing founders of all stripes.
Expertise > Name-Recognition Expertise
There is an argument that educational attainment and work experience correlate with competence. In many cases, I would agree. Undoubtedly, certain institutions attract and mold brilliant minds and provide useful instruction. Still, it takes more than a stellar resume to launch a startup, much less a successful one. So the question is not whether an entrepreneur went to Stanford or worked on Wall Street — though those experiences, at times, can certainly help. Rather, the question is whether an entrepreneur has the expertise consistent with the problem they are solving. Do they understand their product, their competition, their markets? Can they articulate a credible vision that reasonably suggests they can sufficiently win their market? Can they inspire, lead, and execute? Expertise is expertise. What matters is that you know your stuff, not where you learned it.
Entrepreneurship is a battle… find someone who’s battle-ready…
Entrepreneurship is grueling. It’s lonely. Few can do it. Fewer can do it well. Frankly, it’s a battle of marathon proportions. It should follow that it’s not enough to invest in smart people, though that should be mandatory. Investors must also prioritize the presence of grit in founders. What good is a founder’s genius if a little bit of adversity is sufficient to knock them out for the count. Instead, both qualities are required — though perhaps the presence of grit should be factored just a bit higher than raw smarts/expertise, for much of literature supports that.
Final Thoughts
This piece is not about bashing venture capitalists. Rather, the intent is to highlight ways to improve the ecosystem. Pattern-matching is neither good nor bad. What matters is how and for whom the practice is applied. Allowing for new backgrounds to “fit” the “pattern” is a good thing. It introduces fresh ideas and perspectives, which only strengthens the industry, an outcome that we should all support and hope for.
