Table of Contents
The software-as-a-service (SaaS) landscape is experiencing a fundamental shift. For years, SaaS companies were viewed as reliable annuity streams, akin to insurance companies with predictable revenue and enduring profit pools. However, the rise of generative AI has introduced a new level of volatility, causing public market investors to question the long-term terminal value of traditional software models. Lucas Swisher, a growth investor and General Partner at Coatue, argues that we are moving away from the era of "good" SaaS companies toward a focus on "platform" companies that can capture disproportionate value in an AI-driven economy.
Key Takeaways
- SaaS Terminal Value is Under Scrutiny: Public markets are devaluing SaaS companies because AI models (like those from OpenAI and Anthropic) make it harder to predict which software moats will remain durable.
- The "Platform Company" Advantage: Investors should prioritize companies that can hop multiple "S-curves" and expand their total addressable market (TAM) through constant product reinvention.
- Margins vs. Scale: Early-stage gross margins can be misleading indicators during architectural shifts; terminal operating margins are the more critical metric for long-term health.
- The Future of AI is Labor Replacement: AI is transitioning from a "copilot" or assistant role to a "labor" role, potentially creating outcomes significantly larger than those seen in the SaaS era.
- Valuation as a Secondary Concern: For generational companies growing exponentially, entry price is often the least important factor compared to market size and founder quality.
The Fragmentation of Public and Private Software Markets
The boundary between public and private software valuations is currently under immense pressure. Public SaaS companies are facing significant headwinds as investors grapple with the implications of AI on legacy business models. Swisher notes that for the first time in the history of the cloud, the "annuity stream" nature of software is being called into question. If an AI model can perform the task of a specialized software tool, the durability of that tool's revenue becomes suspect.
This uncertainty has led to a flight to quality. While public markets offer liquidity, Swisher argues they often struggle to capture the "future." Most high-growth, generational AI companies—such as OpenAI, Anthropic, and SpaceX—remain private. These platform companies are growing at scales and speeds that were previously unseen, often choosing to stay private longer to execute on multi-product strategies without the scrutiny of quarterly public earnings.
"For the first time ever with this AI wave, people are questioning the terminal value of SAS."
Defining the "Platform Company" in the AI Era
In the previous generation of software, the "triple, triple, double, double, double" growth metric was the gold standard. In the current era, Swisher believes this framework is insufficient. He categorizes the most successful investments as "platform companies"—entities that do not just dominate a single niche but show the ability to skip between different markets and reinvent themselves. Databricks and Canva serve as prime examples of this phenomenon.
The Ability to Hop S-Curves
A true platform company demonstrates the capacity to ride one wave of technology and successfully transition to the next. Databricks evolved from a data transformation layer into a center for model training and enterprise data management. Similarly, Canva transitioned from a niche yearbook tool into a comprehensive design suite that integrated AI long before it became a mainstream requirement. These companies do not just grow; they expand their TAM through continuous product evolution.
Market Pull and High-Growth Dynamics
When a product truly resonates in the current market, it is "yanked" into the enterprise at a rate that exceeds traditional sales cycles. This "market pull" is a stronger indicator of success than any specific sales metric. If a company's revenue is screaming rather than just growing, it suggests a level of product-market fit that can justify high entry valuations.
Reevaluating Valuation and the Power of the Double Down
For growth investors, the entry price is frequently a point of contention. However, Swisher maintains a counterintuitive stance: price matters, but it matters least. When a company is growing 10x or 50x year-over-year, a valuation that seems "insane" today can look exceptionally cheap within twelve months. The primary focus should be on the scale of the "big idea" and the potential for the company to become a multi-hundred-billion-dollar entity.
"I think price does matter but I think it matters least. Margin matters but early it can be a misleading indicator."
One of the core strategies at Coatue involves the "double down" round. Swisher highlights that as companies move up in market cap bands, the percentage of companies that successfully 10x their valuation actually increases. This makes the ability to reinvest in winners more valuable than a "spray and pray" approach at the seed stage. By concentrating capital in the top 20 platform companies that generate 80% of the total value in the ecosystem, growth firms can secure more reliable, outsized returns.
Data as a Prerequisite, Not the Answer
Under the mentorship of industry legends like Mary Meeker and Mamoon Hamid, Swisher learned the importance of analytical rigor. However, he cautions against "missing the forest for the trees" by living exclusively in Excel. Data is a tool for storytelling and identifying inflection points, but it cannot replace the qualitative assessment of a founder's vision or a market's underlying momentum.
"Data is a prerequisite. It is not the answer. The data must be very good, but it's not the whole picture."
In the context of AI native businesses, data helps identify fragility. For instance, a low-margin AI company must demonstrate exceptionally high customer retention to survive. If the customer behavior isn't "sticky," the business model becomes incredibly fragile because there is no margin for error. Investors must use data to confirm that customer behavior supports the long-term thesis, even if current financials look unconventional due to high inference costs.
The Shift from Assistant to Agentic Labor
The most significant change in the last twelve months is the realization that AI is moving beyond simple assistance. We are entering a period where machine inputs are replacing human inputs, effectively turning AI into a "token factory" for labor. This shift suggests that the total addressable markets for AI companies are much larger than the software markets of the past because they are capturing portions of the global labor spend.
Accelerated Adoption Cycles
The speed of AI adoption is outpacing the SaaS era. When the three major hyperscalers (AWS, Azure, Google Cloud) were at a $9 billion run rate, they were growing at approximately 60%. In contrast, current AI leaders are seeing growth rates near 800% at similar scales. While enterprise change is historically slow, the "text-in, text-out" nature of AI allows for faster integration than previous industrial or agricultural revolutions.
The Advantage of Optionality
Strategic positioning is as important as technical superiority. Swisher notes that companies like Anthropic have gained an advantage by building for every cloud and every chip platform. In a world where compute capacity is often constrained, the ability to deploy across various infrastructure providers makes a company more resilient and cost-effective. This "multi-cloud" approach creates a network of partners who are invested in the company's success.
Conclusion
The current technology cycle is fundamentally different from the SaaS wave that preceded it. Investors and founders must adapt to a world where terminal value is questioned, margins are nuanced, and the potential for labor displacement creates massive new pools of revenue. Success in this environment requires a focus on platform companies that can reinvent themselves, a willingness to ignore short-term valuation "noise" in favor of big ideas, and the discipline to double down on the few companies that will define the next decade of technology.
