SaaStr 843: Software Stocks Have Massively Crashed. Here's What Founders Need to Know.
Summary
The episode examines the current SaaS and VC landscape through the lens of AI adoption, product growth, and event economics. Jason Lemkin argues that merely adding AI features is insufficient—true AI companies must show re-accelerated growth driven by agents or integrations. He explains how AI agents are already replacing human capacity, driving revenue (including an agent that closed a $100K deal) and reshaping discoverability and vendor selection. The conversation also covers the high fixed costs of large conferences, why private equity is shying away from many legacy B2B businesses, and how low-friction “vibe-coding” tools are flooding niche markets with clone-ready apps.
Key Takeaways
- 1AI must re-accelerate growth to matter — productizing AI features alone isn’t enough.
- 2AI agents can replace human work and directly generate revenue.
- 3AI discoverability will reshape how businesses choose software.
- 4Vibe-coding and low-friction tooling are flooding niches, lowering differentiation.
- 5Running large-scale events requires a very large upfront investment but can be highly profitable at scale.
Notable Quotes
"My simple rule guys: If growth is not accelerating, you're not an AI company."
"It's going to cost you $10 to $15 million, maybe even 20 to turn on the lights."
"We had an agent that closed a $100K deal on Saturday night."
"When people do discovery, they're going to ask their agent."
Episode questions
How should founders judge whether their AI work actually matters?
Measure whether AI integration re-accelerates growth or improves core revenue metrics; if it's just a demo or feature without growth lift, it won't move the needle for investors or markets.
What tangible benefits did agents bring to the speakers' business?
Agents handled reactivations and follow-ups with a 70% open rate, reduced headcount needs (2–3 people doing the work of 15), and even closed a $100K deal autonomously, proving both efficiency and direct revenue impact.
Are events still a viable business for media/tech companies?
Yes, but events require large upfront investment ($10–20M 'nut') and significant focus; they can be extremely profitable once scaled, but are risky and distract from core business if pursued marginally.
How will AI change software discovery and vendor selection?
Agents will become the discovery layer: users will ask agents which product to use and agents will recommend vendors, meaning companies must optimize for agent discoverability and trust to capture demand.