Most important take away

More than $2 trillion in market cap has been wiped from SaaS businesses in Q1 2026, but much of the sell-off is driven by fear rather than actual financial deterioration — many beaten-down SaaS companies are still posting all-time-high revenue and free cash flow. The key differentiator for investors is the revenue model: pay-per-seat SaaS companies face the greatest structural risk from AI, while cybersecurity and platform companies with usage-based or mission-critical offerings are better positioned to survive or even thrive.

Summary

Stocks and Investments Mentioned

• iShares Expanded Tech Software Sector ETF (IGV): Down over 30% in the last six months vs. NASDAQ down only 9%. Demonstrates the broad investor exodus from SaaS.
• Chegg (CHGG): Down 99% from 2021 peak. Revenue falling 40% year-over-year. The most extreme real-world example of AI disruption destroying a SaaS business.
• ServiceNow: Stock down over 50% in the past year despite subscription revenue growing 20%+. Jensen Huang specifically named it as a company AI agents will use rather than replace.
• Datadog: Down nearly 40% from highs despite bookings surging 37% year-over-year. A clear disconnect between stock performance and business results.
• Adobe: Trading at roughly one-third its historic price-to-earnings valuation.
• HubSpot: Trading at its lowest valuation ever (4x sales) since going public, yet trailing 12-month revenue and free cash flow are both at all-time highs. The hosts were uncertain about its future, recommending staying on the sidelines.
• Constellation Software: Trading at 3x sales, cheapest since the Great Financial Crisis, with revenue and free cash flow at all-time highs. Same cautious stance as HubSpot.
• Asana (ASAN): Identified as highly vulnerable. Smaller than competitors with inferior growth rates. Net retention rates have dropped below 100% (existing customers spending less). Its pay-per-seat model is particularly exposed as AI makes teams more efficient and reduces headcount.
• Atlassian (TEAM): Identified as highly vulnerable. Despite 23% year-over-year revenue growth, it showed its first-ever decline in enterprise seat counts. Its JIRA and Confluence products face the same per-seat pricing risk since AI agents cannot be charged seat licenses.
• CrowdStrike (CRWD): Down 30% from recent highs but identified as a strong survivor. AI creates new cybersecurity threats (malicious prompts, AI-generated malware) that require AI-native security solutions. Revenue growth expected to accelerate. Used by the majority of the S&P 500.
• Zscaler (ZS): Down 60% from recent highs but well-positioned. Its Zero Trust access model becomes more critical with agentic AI. Management estimates securing agentic AI operations is a $19 billion untapped market opportunity.
• Autodesk: Briefly mentioned as a safe pick since construction workers are unlikely to abandon CAD software.
• Duolingo (DUOL): Down 80% from its 52-week high, now trading at 4x sales (down from 32x). Identified as a controversial survivor pick due to strong branding, network effects, and the fact that it is free to use — making a competing AI-built app unnecessary for most users. Management is focusing on the free tier, leading to lower near-term bookings but still forecasting strong three-year growth.

Actionable Insights

1. Avoid pay-per-seat SaaS companies in tech-facing markets. The per-seat revenue model is structurally threatened by AI. As companies become more efficient with fewer employees and AI agents that do not require seat licenses, revenue will shrink even if the product remains useful. Asana and Atlassian are the clearest examples of this risk.

2. Look for the disconnect between business fundamentals and stock price. Many SaaS stocks have been sold off on fear alone. Companies like ServiceNow, Datadog, HubSpot, and Constellation Software are posting strong or record financial results despite massive stock declines. If AI disruption does not materialize for these companies, current prices represent a significant contrarian opportunity.

3. Favor cybersecurity SaaS. CrowdStrike and Zscaler are positioned to benefit from AI rather than be disrupted by it. AI creates entirely new threat vectors that demand AI-native security solutions. Both are down significantly from highs, potentially offering attractive entry points.

4. Watch for revenue model evolution. The hosts predict SaaS companies may shift from per-seat to usage-based or token-based pricing models (similar to how Claude charges per token). Companies that successfully make this transition could preserve revenue even as AI agents replace human users.

5. Use Chegg as a cautionary benchmark. Chegg shows what happens when AI directly replaces a core product with free alternatives. When evaluating any SaaS holding, ask whether the product can be fully replicated by freely available AI tools. If yes, the risk is existential.

6. Consider branding and network effects as moats. The Duolingo thesis rests on the idea that even if AI can replicate the product, switching costs, brand loyalty, and network effects protect the incumbent. Apply this lens to other SaaS holdings.

Chapter Summaries

Introduction and Setting the Stage: Tyler Crowe introduces the episode as a deep-dive responding to listener Scott Pounder’s question about SaaS vulnerability to AI. John Quast explains why SaaS companies have historically been investor favorites (high margins, recurring revenue) and why AI now threatens them — a few skilled AI prompters can build competing software in days. The IGV ETF is down 30% in six months.

Early Signs of Disruption: Matt Frankel walks through Chegg as the most extreme example of AI disruption (down 99%). He notes that for most SaaS companies, the sell-off is driven by investor fear rather than actual business deterioration, with an estimated $2 trillion in SaaS market cap erased in Q1 2026 alone.

Best and Worst Case Scenarios: The worst case is AI rendering SaaS businesses far less useful, with customers coding their own solutions or needing fewer seats. The best case, supported by NVIDIA CEO Jensen Huang, is that AI agents will use existing enterprise software tools rather than replace them.

HubSpot and Constellation Software: Directly answering the listener’s question, John notes both companies have all-time-high revenue and free cash flow but historically low valuations. The verdict: uncertain — if investors are wrong about the doom, these are great contrarian buys, but the hosts would personally stay on the sidelines.

Most Vulnerable Companies: John flags Asana as a weak, small player already seeing net retention drop below 100%. Matt identifies Atlassian, noting its first-ever decline in enterprise seat counts despite 23% revenue growth. Both companies share the pay-per-seat model, which the hosts see as the most exposed revenue structure.

Revenue Model Discussion: The hosts emphasize that not all SaaS is equal. Pay-per-seat models are most at risk. They predict a potential industry shift toward usage-based or token-based pricing to reflect actual product consumption by AI agents.

Survivors and Thrivers: Matt makes the case for CrowdStrike and Zscaler as companies that will thrive because AI creates new cybersecurity threats and expands their addressable market. John makes a controversial pick of Duolingo, arguing its brand, network effects, and free tier make it resilient despite an 80% stock decline.