A VC Shares His Secrets for Picking Investments

Editor’s note: Chris Paik is an investor at Pace Capital, an early-stage venture capital firm based in New York City with $400 million under management. He originally published his theses for evaluating businesses and business models—what he called “frameworks”—as a Google Doc, which he posted on Twitter for others to read and comment on.

It is rare to see an investor publish (let alone open-source) the thinking behind his investment decisions, and we thought that his insights would be valued by a wider audience. We hope that these ideas are useful to you—let us know what you think in the comments.

I believe there is a science to venture capital—a “business physics,” if you will. There is a lot written about the science of investing and optimal risk, but none of the existing literature has fully explored the limits of the science of venture. It’s easy to look at someone who says they want to run a lemonade stand and tell them it’s never going to be a venture-backed business. If we can be so confident in that “no,” then what are the limits of that confidence? 

We may never have business theories that are as reliable and precise as scientific theories—but we can develop frameworks that help us think about business more rigorously. For example, a framework I use is to think about catalyzing moments in technology—i.e., to ask, “Why now?” The vast majority of capturable enterprise value happens right after a catalyzing moment. It’s why we haven’t seen a single new mobile social application since 2014, with the launch of Musical.ly (now TikTok). The “Why now?” that supported Twitter, Snapchat, Instagram, and TikTok was the distribution of cellular bandwidth speeds. Twitter came first, because it was initially only text, which has the lowest wireless packet size. Then came Instagram with images, Snapchat with images and video, and TikTok with video. That couldn’t have happened in any other order. If there were any other mobile media platforms, we would have seen them emerge in line with that spectrum of cellular bandwidth deployment. 

I use these frameworks in the same way as I experience a visit to the optometrist: you sit in the chair and stare at fuzzy letters, and they start to come into focus. Often I’ll layer them: there will be one core framework, and then I’ll try different combinations. Sometimes it’ll help me see more clearly, or sometimes it’s not right. Sometimes there will need to be a hand-whittled lens that will bring me clarity at the end—and that’s because it’s something I’ve never seen before. I use them to get a base-level understanding of a new opportunity. 

This piece is the fundamentals—the stuff that can be known. Many of my thoughts draw on and synthesize concepts from economics, math, physics, chemistry, biology, and psychology. I rely on a number of core concepts as inputs, including behavioral economics, cognitive biases, entropy, game theory, moral hazard, nudge theory, potential and kinetic energy, price elasticity of demand, social psychology, supply and demand, and thermodynamics, along with many others. I can only assume that there are investors who disagree with some of these frameworks, which I welcome—it took 1,000 years of experiments and debate to arrive at the heliocentric model of the universe.

Let’s dive into the frameworks.

Atomic value swaps

An atomic value swap is the measurement of the sustainability of repeated core transactions in an ecosystem. Payment for goods or services is an example of an atomic value swap, one where cash is exchanged for an item or service. Both parties deem the transaction to be beneficial and therefore the transaction occurs. If the price of a product or service is too high, the buyer will not engage in the transaction.

Secular-to-secular (money/goods/services) atomic value swaps are easily understood by the market clearing price of the exchange. Secular-to-sacred atomic value swaps (in which the sacred is uniquely priced to the individual) are significantly more complicated to understand, as behavioral psychology tends to skew expected reactions from supply or demand to changes in the transaction.

This extends best to less tangible exchanges as a concept to understand non-monetary transactions. For example, asking for an invitation to a new product is exchanging fractional social indebtedness for a scarce resource.

The three questions that help me ring-fence the atomic value swap in any situation are:

• What is the value being delivered?
• What is the perceived value of what is delivered?
• How fairly compensated is the creator for value delivered?

Examples: Instagram has two main atomic value swaps with its users, one to creators and one to consumers. Creators of content expend time and energy in the form of work in exchange for distribution and building an asset that can be monetized (audience). Consumers are willing to view ads in their consumption experience in exchange for entertainment. A third atomic value swap exists to advertisers, which is a typical secular-to-secular exchange of money spent on advertising (“ROAS,” or return on ad spend).

Smaller atomic value swaps exist at the product feature level. The action of liking a post is a consumer expending a fractional amount of work to essentially “thank” the creator (and a small amount of social signaling, but to a lesser degree). Such an exchange would not occur if the hurdle to thank the creator was too high. By reducing the friction to engage in the interaction low enough, the atomic value swap occurs and is sustainable.

Business model–product fit

“Business model–product fit” is just as critical for a company to succeed as “product–market fit.” Optimal business model–product fit is the company level version of the atomic value swap.

Examples: Marketplaces generally take 20% of every transaction that occurs on the platform. Why has the invisible hand determined that 20% is the correct take rate for marketplaces? It’s important to understand the alternative to marketplaces and what they offer to the supply side, which typically bears the cost of the fee rather than the demand side. The marketplace offers pre-existing demand to incremental supply. A would-be supplier on a marketplace can rest assured that if they put their correctly priced goods or services on a liquid market, demand will show up and transact. It’s not unusual for companies to spend approximately 20% of their top-line revenue on marketing, which is essentially demand generation. If the supplier were to outsource all of their demand generation to a single entity, that entity would be entitled to collect what would otherwise have been spent, or about 20%.

OnlyFans’s innovation was one of business model–product fit. Credit card chargeback rates on adult sites were too high for payment gateways to incur the cost themselves. As a result, many payment gateways categorically refused to service adult content sites. OnlyFans internalized this additional chargeback risk and passed it on to its users in the form of a higher take rate. In a normal environment, a 20% take rate without delivering demand would be unsustainable.

Frameworks within business model-product fit

Low ACV (average contract value) demand cannot require education

If the expected economic return on an acquired customer is low, any acquisition path that requires education of that consumer to the virtues of the product will inevitably lead to failure unless a macro tailwind or zeitgeist eventually eliminates the educational cost. Most self-serve products fall into this bucket.

High ACV demand must require education

This is enterprise sales in a nutshell. If the addressable demand for a product with high ACV did not need to be educated about the virtues of the product, market forces would naturally give way to a competitor that charges a lower ACV and acquires customers more effectively. If a high ACV product “sells itself,” the producer should anticipate lower-cost competitors to enter the market.

User-generated content > editorially driven content for ad-driven platforms

It is impossible to compete for the same ad dollars with a product that has strictly higher COGS (cost of goods sold).

Examples: The ongoing crisis of how to sustain the business of journalism in a post-internet world was born when news moved online and began competing for the same advertising dollars as UGC platforms. Prior to the shift, news publications were very attractive businesses with a large market share in advertising and a healthy paid consumption model. See the New York Times’s revenue collapse and slowly recover through the focus on subscription.

Editorially driven content > user-driven content for subscription-driven platforms

No user-generated content platform that offers distribution (and not monetization) to the marginal content creator can succeed with a paywall.

Examples: Netflix and Spotify are great examples of paywalled content that primarily offer monetization to their suppliers. Medium is a good example of a failed paywall because its original value proposition to supply was built more around distribution than monetization.

The seven deadly sins are actually the seven core motivators

All successful consumer-facing companies appeal to one or more of the seven deadly sins. They are time-tested core motivators that incentivize people to do things; the fact that they have survived for all of time without any edits is proof of their power. There are no successful consumer companies that do not appeal to any of the seven deadly sins.

Different motivators can apply to different constituents within each company, and even different behaviors from the same constituent.

Examples:

Sloth: Uber, Amazon

Pride: Instagram, TikTok

Gluttony: DoorDash, Netflix

Lust: Tinder, OnlyFans

Envy: Pinterest

Wrath: Twitter/X

Greed: Bitcoin, Robinhood

Sloth tends to be the easiest to monetize because the end user places a fairly consistent price on the trade-off between money and time (convenience). 

Pride is also easy to monetize—see the framework on elasticity of demand and Maslow’s hierarchy of needs. 

Gluttony is straightforward in its monetization as it is rooted in consumption. 

Lust, while easy to monetize, has historically been confused with long-term mate finding, a seemingly impossible atomic value swap that has plagued dating websites. 

Envy is slippery to monetize, as the challenge is how to own the point of purchasing decision (the path from inspiration/envy to a monetizable transaction can be long and convoluted). 

Wrath is a very difficult sin to monetize and often manifests through in-group/out-group dynamics. 

Greed is the hardest of all sins to monetize, naturally so as the user is loath to engage in a sub-optimal transaction, and will prefer to be monetized via any other sin (i.e., sloth in the form of performance fees).

Why ‘Why now?’

Venture capital is a very specific instrument that is purpose-built to fund companies that are capable of explosive value creation over compressed periods of time. Concurrently, the invisible hand is a constant force that continually reduces the ability of any one company to generate outsized value. That means that venture capital is particularly well suited to finance companies that are capitalizing on “dam-breaking” moments—sudden changes in technology and regulation (and to a lesser extent, capital markets and societal shifts). Each time a new shift occurs, it is analogous to the formation of an unstable radioactive isotope. The radioactivity throws off a huge amount of energy in the form of capturable enterprise value, but is subject to half-life decay (the invisible hand’s movement). Over time, the isotope decays and eventually becomes lead, at which point no new companies can generate enterprise value from the shift. Without a sufficient answer to the question of “Why now?” any venture capital invested into the company or category is subsidizing company building that would be better served by alternative capital instruments (with a lower cost of capital, i.e., debt, etc.).

Examples: A helpful metric to examine is “enterprise value per year,” which often elucidates the distinction. Since inception, Facebook has generated an average of $43 billion in enterprise value for every year of its existence (Google is $54 billion per year, Apple is $50 billion per year). Conversely, Disney has generated an average of $3.5 billion in enterprise value for every year of its existence, a full order of magnitude off (Nike is $3.9 billion year, etc.). To put this in perspective, Lululemon, a great example of a successful consumer clothing company founded in 1998, is worth $43 billion—Facebook has created a Lululemon every single year since 2004.

The reason why we have not recently seen additional mobile-first social companies emerge is because we are well into the half-life decay of the smartphone. The vast majority of the capturable enterprise value in the mobile social space was done in the first few years after smartphone saturation (Snapchat was founded in 2011). In the same way, we have not seen a successful new company leverage DVDs, even though at one point it was a watershed moment in technological capability. 

Uber, Lyft, and other ride-hailing companies could not exist pre-smartphone. More than simply a homogenous operating system to connect drivers and riders via the same software, the most important technological enabler for on-demand innovation was cellular networking speeds. The original iPhone was released using 2G/Edge, which was too slow to support real-time GPS and turn-by-turn directions. It wasn’t until the iPhone 3G that cellular bandwidth was capable of delivering these critical features. While the original iPhone was released in 2007, the iPhone 3G took another year, being released on July 11, 2008. Uber was founded nine months later.

Being the answer to ‘Why now?’ for other companies

If a company can deliver, mostly through technological innovation, an answer to the question, “Why now?” for other companies, it will be a venture-scale outcome, assuming proper business model—product fit. The challenging part here is that the vast majority of the customer base for the innovating company does not yet exist at the time of founding (market risk). 

Examples: The classic example is Fairchild Semiconductor. The introduction of semiconductor technology enabled a Cambrian explosion of new businesses to exist. A more modern version of this is Plaid, without whose existence many fintech startups would not be possible (too much consumer friction = unsustainable CAC).

Market risk vs. execution risk

Market risk is where the demand for the product is unknown. Execution risk is where the demand is well understood, but the hard part is in the delivery of value against existing competition. Any company that is pure execution risk without any market risk is not a suitable venture investment.

Examples: Instagram, Snapchat, and most of what we consider “consumer” companies neatly fall into the market-risk bucket. No amount of money spent on a customer survey or consulting project would yield the conclusion that there is an opportunity. It requires an explorer to basically set sail with conviction and strike land.

An example of execution risk would be becoming a franchisee. There is well understood demand for the product, but delivery of that product to demand is not so simple. Opening a new location in a new city or country would be some small amount of market risk, so near pure execution risk would be opening a franchise location a few blocks away from another one.

Incentivizing time speculation is the most powerful engine in value creation

Speculation, defined as taking risk through the allocation of resources, is an incredibly powerful feedback loop. If a platform can correctly incentivize users to speculate with their time and energy, it will be very successful. The speculation of time and energy is an easier atomic value swap to extract value from than the speculation of capital, as the cost of the middleman is more evident in the latter and will promote disintermediation. A common way that this is phrased is, “Build a platform for others to build their own businesses on top of.”

Example: The United States achieved its position as the leading superpower in the world by being the most attractive country for time speculators. It now has a monopoly on ambitious immigrants, which serves as a lucrative bucket fill rate, regardless of leaks. Further, the beginning of the end for the U.S. will be when another country achieves a strictly better value proposition for time speculators either through a more attractive offering, or the erosion in expected value of the American Dream.

All user-generated content networks enfranchise a class of creator that was previously disenfranchised

The reason why Charli D’Amelio was the most ascendant first TikToker is because she is a dancer. TikTok is the first platform to incorporate audio as a native part of the core product consumption experience, and thus the first opportunity for dance to be properly appreciated as content.

JFK would have been less advantaged in his presidential campaign without the invention of the television. Similarly, FDR would have struggled in a post-TV era.

Another way to interpret this framework is that a new content network that aims to poach top users from a pre-existing one will fail. Success stories in every new network will be homegrown—the opportunity/switching cost for would-be emigrants established on other platforms is too high.

Supply as commodity vs. supply as unique are fundamentally different strategies

Treating supply as a commodity is the core philosophy of all marketplaces. This leads to supply competing to serve the firehose of demand with that competitive dynamic translating into a consumer surplus. Treating supply as unique is the classic “arm the rebels” approach, which requires the supply to think of itself as non-fungible. Supply that believes itself not to be a commodity will invest in products that allow themselves to differentiate from competitors, capture more margin from their customers, and avoid being platform dependent.

Examples: Uber is an excellent example of supply as commodity. Interestingly enough, I explicitly wouldn’t want the same driver as I had last time because in all likelihood, they are very far away when I need them. Shopify is a great example of supply as unique. No company using Shopify thinks of itself as a commodity, which is exactly why it invests the time and energy into setting up its own storefront rather than plugging into a marketplace with pre-existing demand.

Potential energy in an ecosystem must be converted into kinetic energy

Every new company addresses pockets of potential energy that has yet to be converted into kinetic energy. The product offering targets these pockets, typically within the end user or customer, and works to remove barriers to allow or catalyze the natural conversion of that potential energy into kinetic energy. Importantly, incentives govern the conversion process and are immutable, like gravity and the laws of physics.

Examples: The sharing economy is built on the idea of identifying pockets of untapped potential energy (latent inventory), and converting it into kinetic energy (merchandised supply with little to no opportunity/carrying cost). By converting this potential energy into kinetic energy, sharing economy companies added new supply to pre-existing exchanges, lowering the market clearing price and monopolizing the new supply. Total addressable market can be conceptualized as the integral of potential energy in an ecosystem.

Elasticity of demand approaches zero as you ascent Maslow’s hierarchy of needs

Assuming that items are not subject to scarcity pricing of utility value (i.e., there is a pandemic and toilet paper is out of stock), consumers have an increasing willingness to pay any price as you move up Maslow’s hierarchy of needs. Self-actualization has virtually inelastic demand. Self-expression also has highly inelastic demand. Never underestimate a person’s willingness to pay to close the gap between how the world perceives them and how they perceive themselves.

Examples: Conspicuous consumption and many Veblen goods (for which price increases along with demand) are examples of esteem needs, and therefore very high willingness to pay. Conversely, physiological needs with pure utilitarian value exhibit very high levels of price elasticity.

A parent’s willingness to spend on healthy food for their child elucidates the difference in the intercept point on the hierarchy. For the child, it is purely sustenance, and left to their own devices, they would not necessarily be willing to pay a higher price. For the parent, the wellbeing of their child is part of the parent’s self-actualization, hence the near inelastic demand.

A specialized tool will always beat a generalized tool over time

A Swiss Army knife is very useful when you are space constrained. It is less useful when you need a dedicated screwdriver to assemble a room full of furniture. Similarly, products with a generalized value proposition will inevitably be cannibalized by more specialized competitors. Convenience is the only defense generalized tools have against erosion by specialized tools.

Examples: Famously, Craigslist as a generalized tool has been competed against by more specialized tools in each of the classifieds categories. Similarly, over time, eBay has been cannibalized by competitors who are focused on a specific vertical within eBay (i.e., Poshmark, Bring a Trailer).

Distribution vs. monetization

Platforms focus on offering the supply side of their ecosystem either distribution or monetization. Those that focus on a combination of both will be challenged by the specialized vs. generalized dynamic over time.

Examples: Facebook, Instagram, Twitter (classic), and TikTok (non-live) offer pure distribution. No constituents on the platforms receive any monetary payment for their participation, incentivized only by the prospect of building an audience and engagement. Shopify and Patreon offer pure monetization. No customer of Shopify or Patreon expects to increase their distribution simply by being on the platform. The main value proposition is in the monetization of pre-existing distribution. YouTube and Twitch are good examples of platforms that offer both distribution and monetization. Rather than being the best of both worlds, over time, it is often the case that they offer the worst of both worlds, leading the supply side to multi-home across distribution platforms and turn to more effective monetization tools (an example of the specialized vs. generalized framework). 

A running hypothesis is that by introducing ad monetization to its user base, X is violating its original contract with its users, and it will lead to a fundamentally different kind of content that is economically incentivized. Platforms that try to economically incentivize their users directly are effectively acquiring laborers, and more often than not, people who make a little money will decide that that they don’t want to be a laborer and/or, when it becomes hyper-competitive to make money, lack the motivation to continue their labor. There are plenty of people who joined Uber when it was easy to make money, and when it became hyper-competitive, they left.

10x experience to a single, critical customer

Every company has a singular critical customer, the cornerstone on top of which the business would not exist otherwise. The atomic value swap with the critical customer is the most important exchange to ensure long-term viability. The company can only attain its initial foothold within its critical customer base by offering a 10X better product than the next best alternative. A product that is only 2X or 3X better will not have enough activation energy to overcome the inertia of switching cost.

Examples: Many companies’ critical customers are fairly simple. For example, Shopify’s critical customer is the incremental e-commerce business. Salesforce’s critical customer is the incremental sales division. The critical customer can get hazier with multiple constituents, like a marketplace or user-generated content network. For marketplaces with a single SKU, or where the customer does not actively choose the supplier (Uber), their critical customer is the demand side. For marketplaces with multiple SKUs, or where the customer engages in active choice (AirBnB), their critical customer is the supply side. All user-generated content networks’ critical customer is the content creator.

10X is hard to come by through a single optimization (i.e., a 10X more delicious apple). It is typically achieved by a combination of vectors that multiply together (i.e., 5X cheaper, 2X better = 10X). This is the basis for the common saying “cheaper *and* better.”

Understanding selling picks and shovels 

An oft-cited phrase in venture capital is that the venture play in an ecosystem is to “sell picks and shovels” (another, less popular phrase is to open a laundry business). This draws on the California gold rush in 1848-9, where the only reliable way to make money amidst the boom in speculation was to offer goods or services to the demand. What is often lost in applying the idiom to a new market or opportunity is more deeply understanding the demand side. This is best understood by the disparity in the answers to the questions:

• What was the demand for picks and shovels in 1847?
• What was the demand for picks and shovels in 1849?
• What was the demand for picks and shovels in 1856?

To perfectly apply the idiom to a new market, the demand for the goods or services must be undergoing an insane amount of explosive growth. It is not merely sufficient to offer goods or services to a pool of demand that is growing at a steady pace, as the invisible hand will likely act in lock step and reduce the addressable opportunity. This rhymes with the question of “Why now?” but examines the same time period through a lens of serviceable demand.

Examples: Many companies cater towards a specific generation, be it baby boomers, millennials, or Gen-Z. Unfortunately, the rate of aging of a population is simply not fast enough to warrant explosive growth. More poignantly, there are many companies in the “deathcare” space, aiming to meet demand for end-of-life services. Short of some cataclysmic disease or natural disaster, the death rate of any population simply will not undergo anywhere near the same kind of growth rate that “gold prospectors” did in the 1840s. A true venture scale company cannot be dependent on “time to pass” to naturally deliver it customers.

A contemporary example of this would be Robinhood, whose business model, payment for order flow, is indexed towards the number of people who identify as retail investors. By dropping the cost of trading to zero, Robinhood ushered in a new wave of demand and capitalized on serving it.

First order irrational, second order rational

An effective strategy to unlock potential energy in what may seem to be a calcified ecosystem is to do something that the existing, entrenched players deem to be completely irrational. The conceit in this strategy is that while the behavior may seem irrational at the first-order level, it is rational at the second-order level and often leads to a market leading position if not monopoly.

Examples: Credit Karma is a perfect example of this strategy. When it was founded, the credit bureaus all made very good money by charging for credit reports. Whether by corroboration or complacency, Equifax, Transunion, and Experian neglected to rock the boat and were content in their business model of clipping coupons from customers paying to access their credit reports. By offering credit reports for free, Credit Karma employed a strategy that on face value seemed entirely irrational to the credit bureaus—why ruin a good thing? But Credit Karma was after a much more lucrative pot of gold—financial referrals. Banks and credit cards would pay hand over fist for new customers, and Credit Karma had them in spades because of its strategy. This was the second-order optimization the company was playing for—and it worked beautifully.

Non-P&L operational leverage

A powerful dynamic that leads to steep growth is when a company benefits from contributed work product by non-employees. In question form, this translates to: “How many man-hours of work do you get that you don’t pay for?” This applies widely across categories of work and leads to a dramatically compounding effect over time.

Examples: Most communities are great examples of this. Developer communities help each other troubleshoot, leading to valuable CX (customer experience) leverage.

User-generated YouTube tutorials lower educational burdens and provide sales leverage.

Moderators of subreddits, Discord servers, and Twitch channels all contribute work that none of the companies pay for.

The supply side of marketplaces constantly works to merchandise themselves for the ultimate benefit of the company (AirBnB hosts meticulously detailing their listings, Uber drivers researching highest yielding areas, Etsy sellers merchandising their stores).

All UGC networks exhibit this to an extremely high degree.

Word of mouth can be interpreted as marketing man-hours whose cost is not borne by the company itself.

User trust = consistently meeting expectations

Long-term product success is a function of continually meeting and/or exceeding user expectations. Each time a user engages with a signifier (push notification, search query, email, opens the app, visits the website, makes a purchase), the product builds trust with the user (signified meets/exceeds expectations) or betrays the user (signified does not meet expectations). Over time, the trust account with each user must maintain a positive balance or else the user stops using the product.

Examples: Rating systems are a product innovation to allow for more consistency in setting and meeting user expectations. In addition to the merchandising primitive that increases liquidity, ratings afford the platform better user retention by decreasing the frequency of betraying user trust.

User retention curves can be understood as the ability of a product to build and maintain the balances of trust across user accounts. Churn occurs when the balance of trust in a user account becomes negative.

Hijacking high signal-to-noise modalities

Similar to evolutionary adaptations in mimicry, an effective go-to-market strategy is to hijack signifiers that have already expended the work to establish a high signal-to-noise ratio with their addressable audiences. This is a form of attention arbitrage that must be coupled with utility that consistently meets user expectations; otherwise, its efficacy will asymptotically approach zero.

Examples: Spam exists solely through this strategy. The signal-to-noise ascribed to mail, email, phone calls, text messages, etc. gets hijacked by spammers to achieve an attention arbitrage. Spam can be best understood as mimicry of high signal-to-noise modalities that consistently underwhelm expectations. Without a corrective mechanism, spam leads to fatigue and the devaluation of the original communication medium.

Cold outreach that is well crafted and meets expectations successfully hijacks the high signal-to-noise modality it occurs in. Cold outreach that does not meet expectations is processed by the recipient as spam.

Facebook hijacked the prestige associated with Ivy League colleges in its go-to-market. By appropriating the .edu email addresses of elite institutions, Facebook was able to achieve a brand positioning that allowed it a top-down distribution motion. Coupled with a UGC network, which delivered the long-term utility, the initial arbitrage was translated into a defensible moat rather than vapor.