AI Made the Old Software Moat Crumble
HEY, remember the golden age of SaaS? For 15 years, the classic bargain was simple: engineering teams built intricate cloud tools, and companies paid recurring per-seat fees for access. This model minted fortunes, epitomized by Slack’s $27.7 billion acquisition by Salesforce—a landmark success, not a universally replicable blueprint.
The barrier to entry was the code itself. Building high-quality, functional software required assembling large teams of 20+ skilled engineers, a significant capital and human resource investment that created a powerful, enduring moat. This allowed early movers to dominate markets, charging predictable subscriptions for their digital fortresses.
Then AI arrived, eroding the very foundation of this model. Advanced AI coding tools now drastically reduce the cost and effort of building functional software. Next-gen models like Anthropic Claude Opus variants can compress a 20-person engineering team to 1-2 operators leveraging AI, making code itself a commodity.
This shift means the traditional software moat has crumbled. When anyone can generate competitive software with relative ease, the value no longer resides in the creation of the tool. The old SaaS model, predicated on the scarcity of engineering talent and the friction of development, is irrevocably broken.
Give Away the Tool. Sell the Work.
Greg Isenberg, a sharp internet generalist, offers a compelling antidote: smart SaaS. His thesis? Give away the basic CRM, dashboard, or app for free. This isn't charity; it's an acquisition wedge designed to attract a massive user base, generating proprietary data and network effects—the true moats of the AI era.
This "free tool" strategy reveals precisely what users actually need, not what product teams think they need. The revenue shift moves beyond static per-seat fees. Instead, companies monetize the "work" itself.
This means charging for:
- AI agents that execute complex tasks autonomously
- Human services providing specialized expertise
- Outcome-based pricing, where payment aligns with achieved results
Sequoia Capital Capital’s observation that “trillion-dollar service is the new software” perfectly encapsulates this evolution. The market for accomplishing tasks is vastly larger than the market for merely accessing tools. The old SaaS model is dead; long live SaaS as a service studio.
The New Price Tag Is the Outcome
Pricing now shifts from access to outcome. Consider a calorie-tracking app: basic logging is free, but personalized meal plans from AI agents or weekly check-ins with a human coach carry a charge. The buyer values results—weight loss, better health—not just a login.
This model fundamentally redefines pricing units. Instead of per-seat fees, companies charge for:
- Agent usage (e.g., tokens consumed, tasks completed)
- Completed work (e.g., a fully generated report, a closed lead)
- Recurring human support (e.g., a dedicated account manager)
Buyers, especially SMBs and prosumers, increasingly demand free tools, only paying when the software delivers tangible value. Why pay for a dashboard when an agent can just do the work? This aligns with insights from firms like NFX, which explore this shift in depth, as highlighted in their piece What Comes Next Is Bigger Than SaaS Ever Was - NFX.
Enterprise, however, operates differently. While some per-seat costs may decrease, Fortune 500 companies still pay a premium for security, compliance, deep integrations, and accountable human support. They don't just want the work done; they need guarantees, service-level agreements, and a vendor to hold responsible. This creates a dual market where free tools attract individuals and small businesses, while enterprises prioritize trust and comprehensive solutions.
Enjoying this? Get one like it in your inbox each morning.
one email a day · unsubscribe in two clicks · no third-party tracking
Free Is a Wedge, Not a Business Plan
Greg Isenberg's “smart SaaS” thesis offers a compelling vision, but free is a wedge, not a business plan. Cheap software is easy to copy; AI’s commoditization means a free CRM or dashboard won’t solve customer acquisition costs or guarantee distribution alone. The market quickly replicates free tools, leaving founders with no durable advantage.
True moats extend beyond replicable code. Founders must cultivate:
- Proprietary workflows and data
- Trusted integrations
- Relentless reliability and compliance
- A crystal-clear Ideal Customer Profile (ICP)
As Sequoia Capital Capital noted, "trillion-dollar service is the new software." The real opportunity lies in identifying a painful job-to-be-done for a specific customer. Use a free tool to earn adoption and gather data, but prove customers will pay for measurable agent or service outcomes. The era of charging per seat for a basic tool is over; the future is about delivering undeniable, paid results.
Frequently Asked Questions
Is SaaS really dead?
No. The traditional model of charging mainly for access to a software tool is under pressure; software is evolving toward AI-powered services and outcome-based pricing.
Why might SaaS tools become free?
AI lowers the cost of building software, making free tools a possible way to attract users, gather useful data and create network effects.
How can companies make money if the software is free?
They can charge for AI agent work, completed outcomes, human services, transactions or enterprise-grade security and support.
Does free software automatically create a competitive moat?
No. A defensible business still needs distribution, a valuable workflow, reliable integrations or proprietary data that competitors cannot easily reproduce.