SaaS Is Dead. Long Live SaaS! AI And The End Of The Rationing Of Knowledge Work
Bob Morse argues AI will not kill SaaS; it will lower costs, expand demand, and open software to more knowledge-work outcomes.
Intelligence analysis by GPT-5.4 Mini
The piece pushes back on the idea that AI makes SaaS obsolete. Morse argues that lower software production costs, like past efficiency gains in coal and data centers, can unlock more demand rather than less.
The article says smarter tools can make work cheaper and easier, like a tiny factory that helps many more people. That could mean more people get help from software, not less.
Analysis
Core claim
Bob Morse argues that the current “SaaS is dead” narrative misreads what AI is doing to software economics. Yes, AI can cut the cost of building software and invite more competition. But, he says, lower cost does not automatically mean lower revenue.
Why he thinks demand expands
Morse leans on the Jevons Paradox: when something becomes more efficient and cheaper, people often use much more of it. He applies that pattern to coal, then to data centers, where improved compute density ultimately led to more demand for capacity, not less.
The knowledge-work parallel
He then turns to Financial Engines as a model for AI-like delegation. The company started by advising 401(k) investors, but demand rose sharply once it began managing the accounts directly. For Morse, that is the key lesson: software can move from helping knowledge workers do tasks to delivering the work outcome itself.
He argues that knowledge work is supply constrained because it is expensive, highly trained, and hard to scale. AI can relax that constraint. In his framing, software companies are not just selling tools to workers anymore; they can sell the result of the work, which could reach people and businesses that never had access to a strategist, analyst, lawyer, or financial adviser.
Bottom line
Morse’s conclusion is bullish on software vendors that use AI to improve productivity or provide outcomes directly. The challenge, he says, is learning how to deliver machine-based knowledge work safely and reliably.
Key points
- Morse rejects the view that AI automatically kills SaaS pricing power.
- He argues that efficiency gains can unlock new demand, not just reduce costs.
- He uses coal, data centers, and Financial Engines as examples of that pattern.
- He says AI can let software companies sell knowledge-work outcomes directly.
- He sees the main hurdle as delivering machine-based knowledge work safely and reliably.
If AI really lowers the cost of knowledge work, software companies could reach many more customers than before. Morse says that could turn software from a tool seller into an outcome seller and expand revenue.
The market could still be right that more competition pressures pricing and makes software businesses harder to defend. Morse also notes that delivering machine knowledge work safely and reliably is a major challenge.
