AI Investment Strategy: Valuation, Margin, and Platform Companies
Lucas Swisher of Co2 Ventures analyzes the shift from SaaS annuities to AI platform companies. Key insights cover why margin is a misleading early indicator, the importance of gigantic TAMs, and the strategic advantage of staying private longer.
The Erosion of SaaS Annuities
The traditional SaaS investment thesis, built on predictable annuity-like revenue streams, is collapsing under the weight of AI disruption. Public SaaS companies face a dual crisis: the terminal value of their software is being questioned as AI models automate core functions, and investors lack clarity on which specific companies will survive. This uncertainty has led to a broad retreat from the sector, as capital flows toward assets with clearer growth trajectories or tangible AI integration.
The Rise of Platform Companies
The investment landscape is shifting toward "platform companies" that stay private longer and scale to unprecedented sizes. These entities, such as OpenAI, Anthropic, and Canva, are generating disproportionate value, with a small number of companies accounting for the majority of private market enterprise value. For investors, this creates a bifurcated market: public markets offer liquidity but limited access to the future, while private markets offer access to the highest-growth assets but with reduced liquidity and higher entry barriers.
Strategic Frameworks for AI Investing
Lucas Swisher of Co2 Ventures outlines a rigorous framework for navigating this new environment. First, margin is a misleading indicator in the early stages of AI companies due to high inference costs; instead, investors should focus on retention and usage. Second, the "big idea" test has evolved: companies must address gigantic TAMs, potentially replacing human labor with tokens, to justify high valuations. Third, valuation is the last question to ask; the primary focus should be on the exponential growth curve and the ability to double down on winners. Finally, data is a prerequisite, not the answer; investors must avoid missing the forest for the trees by combining metrics with qualitative insights on founder adaptability and market dynamics.
Conclusion
The future of venture capital lies in concentrated bets on platform companies that can reinvent themselves across multiple S-curves. Investors must adapt their frameworks to account for lower early margins, larger outcome sizes, and the structural shift toward private markets. Success will depend on identifying companies with the talent density and strategic agility to navigate rapid technological shifts, rather than relying on traditional SaaS metrics.
Key insights
-
Early-stage AI companies often have low or negative margins due to inference costs, which is a normal part of the architecture shift. Investors should not penalize companies for this, as margin profiles are expected to improve as token costs decrease and models are optimized.
Impact: Prevents investors from missing high-growth AI opportunities due to outdated SaaS margin expectations, allowing for earlier entry into winning platforms.
-
The "big idea" test now requires companies to address gigantic TAMs, potentially replacing human labor with tokens. Niche verticals with limited expansion potential are no longer sufficient to justify high valuations in the AI era.
Impact: Focuses capital on companies with the potential to generate $50B+ in revenue, ensuring that high entry prices are backed by massive market pull.
-
Valuation is the last question to ask for exponential growth companies. The primary focus should be on the revenue trajectory and the ability to achieve a 3x return in the next fundraising round, rather than the initial entry price.
Impact: Allows investors to secure positions in generational companies at higher prices if the growth curve supports the return, avoiding the mistake of passing on winners due to price sensitivity.
-
Top AI companies are staying private longer, creating a liquidity gap for public investors. Private markets offer the only access to the highest-growth, durable platform companies, while public markets offer liquidity but limited access to the future.
Impact: Highlights the strategic advantage of private market exposure for capturing the highest returns, while acknowledging the trade-off of reduced liquidity.
-
Data is a prerequisite, not the answer. Metrics like net new ARR are guideposts, but investors must combine quantitative data with qualitative assessment of founder adaptability and market dynamics to avoid missing the broader trend.
Impact: Reduces the risk of making decisions based solely on short-term metrics, ensuring that investors capture the full potential of companies that are reinventing themselves.
Action items
-
Re-evaluate early-stage AI investments by prioritizing retention and usage signals over margin profiles. Implement a framework that accounts for the expected improvement in margins as token costs decrease.
Impact: Enables earlier entry into high-growth AI companies that may be penalized by traditional SaaS margin expectations, capturing upside as margins normalize.
-
Apply the "gigantic TAM" test to all new investments, ensuring that companies are addressing markets large enough to support $50B+ in revenue. Avoid investing in niche verticals with limited expansion potential.
Impact: Focuses capital on companies with the potential to generate disproportionate returns, reducing the risk of investing in businesses that cannot scale to justify high valuations.
-
Shift the valuation assessment to the end of the investment process, focusing first on the revenue trajectory and the ability to achieve a 3x return in the next round. Use entry price as a secondary consideration.
Impact: Allows investors to secure positions in generational companies at higher prices if the growth curve supports the return, avoiding the mistake of passing on winners due to price sensitivity.
-
Increase exposure to private markets to capture access to top AI platform companies that are staying private longer. Develop strategies for managing liquidity risk in private holdings.
Impact: Provides access to the highest-growth assets in the market, while acknowledging the trade-off of reduced liquidity and the need for longer investment horizons.
-
Combine quantitative data with qualitative assessment of founder adaptability and market dynamics. Avoid making decisions based solely on short-term metrics, and focus on the company's ability to reinvent itself across multiple S-curves.
Impact: Reduces the risk of missing the broader trend by ensuring that investors capture the full potential of companies that are adapting to rapid technological shifts.
Quotes
“Margin matters, but early, it can be a misleading indicator.”
“Data is a prerequisite. It is not the answer.”
“I don't think the king making concept is a real thing.”