From Sniper to System: Rethinking Venture Capital
A Conversation with Professor Pengfei Wang & Investigate VC
Venture capital has traditionally relied on a simple model: small teams of investors placing large bets on a handful of startups they believe could become unicorns.
But as the global startup ecosystem grows, now exceeding 150 million companies worldwide, some investors believe the venture model itself needs reinvention. That question became the focus of a recent discussion at BI Norwegian Business School in Oslo, where Professor Pengfei Wang, whose research explores technological innovation and venture capital, met with the team behind Investigate VC.
Their discussion explored a broader idea:
What if venture capital itself could be designed as a system, not just a set of individual investment decisions?
Network Effects and Venture Returns. Why does Investigate VC focus so strongly on companies with network effects?
Mikael Krogh:
We focus on companies with strong network effects because they have powerful structural
advantages. As more users join, the product becomes more valuable, which attracts even more users and creates a self-reinforcing growth cycle.
A well-known consumer example is Airbnb: more hosts attract more traveler's, which in turn
attracts more hosts.
The same dynamic exists in B2B platforms. A supply chain data platform, for instance, becomes more valuable as more shipping companies, ports, and logistics providers connect to it. Each participant adds data that improves the network for everyone.
These are the types of companies we look for early, where network effects can become the engine that drives long-term growth.
Professor Pengfei Wang:
The thesis is intuitive but also academically interesting. If companies with network effects
consistently produce higher returns, we might expect investors to crowd into these opportunities and drive returns down.
But that does not necessarily happen.
One explanation is information asymmetry, not all investors are able to identify these
opportunities early. Another is uncertainty. Network-effect companies may have higher average returns, but the variance of outcomes is also much higher.
Many investors prefer more predictable investments.
From “Sniper” to “Machine Gun” Venture Capital. How does Investigate VC’s portfolio strategy differ from traditional venture capital?
Mikael Krogh:
Traditional venture capital often relies on selecting a small number of startups and betting heavily on them. Our approach is different. If network-effect companies have structurally higher returns, the challenge becomes capturing that return profile systematically.
That means building larger portfolios of early-stage companies so the law of large numbers works in our favor.
Professor Pengfei Wang:
That is an interesting shift. Traditional venture capital could be described as a “sniper
strategy”, targeting a few carefully selected companies. Your model resembles more of a “machine gun” strategy, spreading investments across many startups with similar structural characteristics.
From a theoretical perspective, this makes sense if the underlying asset class (network-effect companies) has a higher expected return.
But it also means the ecosystem around those startups becomes critical. Research suggests VC-backed startup success depends roughly two-thirds on selection and one-third on nurturing, so strong partner networks remain very important.
Building a Venture Operating System. What role does data and experimentation play in your model?
Mogens Mathiesen:
Many startups fail not because the idea is wrong, but because they scale too early or make the wrong commercial decisions. We are building what we call a venture operating system, infrastructure that helps founders experiment faster and make better decisions.
This includes systems that help startups run structured experiments, test product–market fit earlier, and analyse growth signals before scaling.
Professor Pengfei Wang:
That aligns closely with recent research in entrepreneurship.
There is increasing evidence that experimentation improves entrepreneurial performance. But experimentation requires structured data collection and analysis, which many startups do not have.
Operating systems that help founders experiment systematically could significantly improve outcomes.
Mapping Industries with AI: The Sector Twin. Investigate VC is developing what you call the “Sector Twin.” What is the idea behind it?
Mikael Krogh:
The Sector Twin uses AI to map industries and identify where startups can create the most value. By understanding both industry challenges and emerging technologies, the system can help startups navigate the scaling process. It can also connect industry problems with startups capable of solving them.
Professor Pengfei Wang:
This is a very ambitious idea. It reminds me somewhat of earlier crowdsourcing platforms that connected companies with external innovators. But your concept goes further by focusing on entrepreneurship and potentially automating the matching process. If such a system can map industry dynamics effectively, it could provide valuable strategic guidance to startups during the scaling phase. Meanwhile, it is crucial for Sector Twin to take a dynamic approach, as all other market actors are making simultaneous moves. This is related to a crucial topic in academia: How can AI do strategy?
The R&D Twin: Mapping Technology Itself. How does the R&D Twin complement this system?
Mogens Mathiesen:
While the Sector Twin maps industries, the R&D Twin maps technologies, including research
developments, patents, and technological capabilities.
The goal is to better understand how technologies evolve and how they can be commercialised.
Professor Pengfei Wang:
That opens several interesting possibilities.
Technology markets often struggle with valuation questions, for example, how to price patents or assess the value of emerging technologies. A system that maps technological development and knowledge flows could help companies, research institutions, and policymakers make more informed decisions.
Innovating Venture Capital Itself. Is the ambition simply to build a better VC fund – or something bigger?
Mikael Krogh:
The ambition is bigger.
Today venture capital still relies heavily on intuition and networks to find opportunities, despite the enormous scale of the startup ecosystem. We believe venture capital needs new infrastructure -- systems that help discover startups earlier, analyse them more systematically, and support their growth more effectively.
Professor Pengfei Wang:
That is what makes this approach interesting.
Your investment model challenges the traditional way venture capital operates. In many ways, you are acting as pioneer entrepreneurs within the VC industry itself.
A New Model for Venture
Investigate VC combines:
Systematic discovery of startups
Focus on network-effect companies
Large-scale portfolio construction
AI-driven decision infrastructure
Venture operating systems that support scaling
The model builds on the firm’s earlier results, including an origin fund that achieved 28% net IRR and a 5× MOIC.
For Professor Wang, the significance lies not only in the investment thesis, but in the broader ambition.
“Your investment approach really disrupts the traditional way of venture capital.”
