Building the Rails: How VC Firms Scale to Fund the Next S-Curve

The central challenge for today's venture capital firms is a scaling paradox. They must maintain the intimate, high-quality decision-making that defines a successful investment process while deploying capital on a scale that matches the exponential adoption of new technologies. The thesis for navigating this tension is clear: the most successful firms preserve the core decision-making unit while vertically scaling operational capacity.

This principle is rooted in a simple, powerful idea. As Andreessen Horowitz cofounder Ben Horowitz has long argued, an investing team shouldn't be too much bigger than a basketball team. The reasoning is straightforward: the conversation around the investments really needs to be a conversation. A small, tightly-knit group ensures that every voice is heard, that debates are substantive, and that the collective judgment is sharp. This "basketball team" model is the engine for identifying and backing the next paradigm shift.

Yet the capital wave they are trying to ride is unprecedented. In just a few years, the focus has shifted decisively to artificial intelligence. By 2024, AI startups raised a third of all capital in venture funding. This isn't a niche trend; it's the dominant force reshaping the entire investment landscape. The sheer volume of money chasing AI startups-funds flowing from seed to late-stage-creates a massive operational demand that a five-person team cannot meet alone.

This sets up a critical efficiency trade-off. The evidence suggests that the traditional VC model of scaling fund size may not be the path to higher returns. A recent analysis found that recent $1 million-$10 million funds are posting better results than recent funds of more than $100 million. Across multiple performance metrics and fund vintages, the median returns for smaller funds consistently outpaced those of their much larger counterparts. This data challenges the assumption that bigger funds are inherently better, highlighting a potential friction where scale dilutes the quality of decision-making or increases the cost of deployment.

The solution, as exemplified by firms like A16z, is architectural. It's not about abandoning the basketball team, but about building the infrastructure to let it play on a global stage. This means creating specialized verticals, fostering cross-team collaboration, and organizing regular offsites to maintain culture and communication. The goal is to keep the core investment conversation intimate and high-leverage while deploying a massive capital wave through a scaled operational backbone. In the race to fund the next S-curve, the winners will be those who master this vertical integration.

The Infrastructure Layer: Verticalization and Platform Support

The scaling paradox demands more than just a lean core; it requires building an internal infrastructure layer that acts as the rails for exponential deployment. Leading firms like Andreessen Horowitz have solved this by architecting a dual-layer system: a small, high-leverage decision-making unit paired with a scaled operational platform.

The strategy begins with verticalization. To maintain the intimate, conversation-driven process that defines a successful investment, A16z splits its large firm into specialized investment verticals. This allows the core team to stay small-an investing team shouldn't be too much bigger than a basketball team. Each vertical focuses on a specific technology frontier, like AI or crypto, preserving deep expertise and rapid consensus. The firm's growth is thus not measured by a bloated partnership, but by the number of these specialized units it can effectively manage.

This vertical structure is only half the equation. The other half is the platform. A16z invests heavily in internal services that free its partners from operational overhead. As Ben Horowitz notes, platform teams can change what partners do day-to-day. Dedicated recruiting, legal, and administrative teams handle the back-office grind, allowing partners to focus exclusively on identifying companies with exponential adoption curves. This is the essence of scaling the VC product: shifting the partner's time from execution to discovery.

Maintaining communication across these specialized units is critical to avoid information silos. A16z combats this by requiring staff to attend other teams' meetings when investment themes overlap. This cross-pollination ensures that insights from one vertical inform decisions in another, preserving the firm's collective intelligence. Twice a year, the firm also holds offsites with minimal agenda, fostering culture and connection at scale.

The result is a firm that can deploy massive capital while retaining the agility and quality of judgment needed to spot the next paradigm shift. The infrastructure layer-verticals, platform services, and cross-team collaboration-transforms a scaling VC from a collection of individuals into a coordinated machine for funding the future.

Financial Performance and the Exponential Adoption Test

The operational model of verticalization and platform support is ultimately judged by financial outcomes. The evidence shows a market in robust expansion, but also a clear performance chasm. Global venture capital investment surged to $120.7 billion in Q3 2025, marking the fourth consecutive quarter of growth. This capital wave is not evenly distributed; it is being funneled toward the next paradigm shift. By 2024, AI startups raised a third of all capital in venture funding, with companies in the sector raising rounds at unprecedented speed and valuation.

This sets the benchmark for success. The top-tier funds from the late 2010s are the only ones consistently on pace for exemplary returns. A TVPI of 3x is often seen as a threshold for exemplary performance, and the 90th percentile for funds raised in 2017 and 2018 sits at 3.52x and 3.07x respectively. In stark contrast, most recent funds sit well below this mark. For funds raised in 2017, the median net TVPI through Q3 was 1.76x. The model's success, therefore, hinges on identifying companies with exponential adoption curves-like those in AI-that can generate the kind of returns that justify the firm's scale.

The risk, however, is that the very infrastructure built to scale could dilute the firm's core edge. Ben Horowitz has noted that platform teams can change what partners do day-to-day. While this frees partners for discovery, it also introduces a bureaucratic layer. The danger is that the "heat-seeking" investing that attracts founders-the ability to move fast and make bold, conviction-driven bets-gets watered down by process. The firm's own growth could, in theory, make it less agile than the startups it funds.

The bottom line is that this model works only if the operational platform perfectly amplifies the decision-making core. It must allow the firm to deploy massive capital into the AI S-curve without sacrificing the quality of judgment that spots the next exponential leap. The financial data shows the prize is still within reach for the elite, but the path is narrowing. The winners will be those whose internal rails are so well-engineered that they never slow down the search for the next paradigm.

Catalysts and What to Watch

The thesis of efficient VC scaling now faces its first major validation test. The model's success hinges on a few near-term signals that will show whether a firm can deploy massive capital without sacrificing the quality of its bets. The first and most direct metric is the performance of A16z's own recent fund vintages. The benchmark is clear: a TVPI of 3x is often seen as a threshold for exemplary performance. While the top-tier funds from the late 2010s are on pace to meet this, most recent funds sit well below it. Investors will be watching the 2023-2025 vintages closely to see if the verticalized, platform-supported model can finally close this gap and generate returns that justify the firm's scale.

The second signal is the health of the market itself-the AI funding landscape. The model assumes the market can absorb exponential growth. Evidence shows AI startups raised a third of all capital in 2024, but the real test is whether this momentum is sustainable. A key indicator is the median Series B valuation for AI startups. If these valuations continue to climb, it suggests the market's capacity for exponential adoption is intact. A plateau or decline would signal a potential saturation point, challenging the thesis that the next S-curve is still wide open.

Finally, the internal architecture must prove its worth by driving measurable results. The platform services and verticalization strategy should lead to a higher proportion of deals that fit the firm's core "basketball team" size-small, high-leverage investment units. The goal is not just more deals, but a deal flow that aligns with the firm's ability to make rapid, conversation-driven decisions. As Ben Horowitz notes, platform teams can change what partners do day-to-day. The success of the model will be seen if these services free partners to focus on discovery, not execution, thereby increasing the number of deals that can be evaluated with the firm's signature agility. The bottom line is that the rails are built; now the train must deliver.