SaaS is dead?
Greg Isenberg · 2026-10-29
💡 Quick Take
1. Traditional software tools that charge per seat are failing because advanced AI makes writing application code almost effortless.
2. The host advises creators to give away basic tools like CRMs (customer relationship managers) entirely for free.
3. Modern software profits come from selling AI agents that complete actual tasks rather than selling access to static interfaces.
4. According to the host, pricing models are moving away from monthly subscriptions toward token usage and outcome-based fees.
5. Enterprise giants will still pay per seat for security, but software for small businesses is racing toward zero cost.
📊 Detailed Explanation
Slack once sold for twenty-nine billion dollars by putting smart engineers in a room and charging monthly user fees. The host explains that this old model, dubbed dumb software, no longer works today. Advanced AI models allow one person to build what once required twenty engineers. Because code creation is cheap, customer willingness to pay for simple dashboards has collapsed. The host advises software builders to skip the old per-seat model entirely, as users no longer want to pay for raw tools.
In this new landscape, the host recommends releasing core software for free to gather data and build user networks. A health brand, for example, can offer a free calorie counter to build an audience. Money is made later by upselling automated AI agents or real human coaches. Instead of charging thirty dollars per user, founders bill per completed result or per token (a basic unit of text processed by an AI). The host highlights that paying for finished work creates a far bigger market than selling blank toolkits.
Customer scale changes how this strategy applies. Small business owners and everyday consumers will expect software to be free. Enterprise clients from large corporations will still pay traditional seat fees for corporate security and peace of mind. The host points out that enterprise seat prices will still drop over time. Software creators should avoid building basic utilities and focus on niche hubs that sell automated labor.
🎯 Tech Expert Opinion
The host correctly identifies how rapid AI development erodes the pricing power of standard software. When a single developer can generate a functional dashboard in hours, software becomes a commodity. Customers do not want to click buttons and enter data themselves anymore. Selling completed tasks through automated agents directly addresses buyer fatigue with endless monthly subscriptions.
Giving software away for free brings serious financial challenges that founders must watch carefully. Running AI models consumes server compute power that quickly burns through startup cash. If an outcome-based pricing model is calculated poorly, customer queries can cost more than the revenue they generate. Gathering customer data only creates a defensible business if a team reaches substantial scale before their runway ends.
According to the host, this agent-driven model makes sense for solo founders and lean product studios targeting small businesses. Creators can target niche industries, replace manual tasks with AI, and charge for real outcomes. Teams targeting Fortune 500 accounts should wait before discarding traditional seat tiers. Those enterprise buyers still require formal vendor contracts, predictable software budgets, and strict compliance guarantees.
Kanal: Greg Isenberg