SaaS is minting millionaires again (here's how)

Key Concepts

• Sub-niche Focus: Targeting specific, smaller segments within large markets to avoid direct competition with venture-backed giants.
• Workflow Mapping: The process of documenting every step of a business operation to identify automation opportunities.
• Agentic Workflows: Using AI agents to perform mechanical, repetitive tasks rather than just providing information.
• Orchestration Layer: The critical infrastructure that coordinates multiple AI agents, manages retries, and validates outputs.
• Outcome-Based Pricing: Shifting from "per-seat" SaaS pricing to "per-task" or "outcome-based" pricing, which aligns better with AI-driven value.
• Media-First Strategy: Building an audience and brand through content creation alongside the software product.

1. The 30-Step Framework for Future SaaS

Greg Eisenberg outlines a transition from traditional SaaS to an "Agentic" model. The core philosophy is that software is evolving from a tool that humans use to a service that executes tasks autonomously.

Phase 1: Market Selection and Research (Steps 1–5)

• Find a Sub-niche: Avoid massive markets (e.g., general "Finance"). Instead, target specific movements (e.g., "FIRE" for Gen Z). Use tools like ideabrowser.com to identify these.
• Map the Workflow: Document a business’s end-to-end workflow (e.g., a roofing company’s process from lead generation to site visit and quoting).
• Identify Financial Friction: Pinpoint exactly where money changes hands (deposits, payments, material orders). This is where software can insert itself as a "wedge."
• Quantify Value: Identify repetitive mechanical tasks. If you save a business owner 100 hours a year, and their time is worth $400/hour, you can quantify the exact ROI of your software.

Phase 2: Distribution and Audience Building (Steps 6–10)

• Media-First Approach: Don't just build a product; build a media company. Create "scroll-stopping" content on one primary channel (Instagram, TikTok, or X).
• AI-Assisted Content: Use AI (Claude, ChatGPT) as a CMO to generate viral ideas, scripts, and calendars.
• Organic to Paid Pipeline: Study organic posts that gain traction (saves, shares, DMs) and convert those proven winners into paid ads.
• Email Foundation: Capture emails from day one to ensure you own your audience, independent of social media algorithms.

Phase 3: Execution and Automation (Steps 11–18)

• Manual Fulfillment: Start by performing the service manually. This builds deep domain expertise and ensures you understand the "judgment" vs. "mechanical" aspects of the task.
• Agentic Workflows: Separate judgment (human-led) from mechanical tasks (AI-led). Build agents to handle the mechanical steps.
• Tool Integration: Connect agents to real-world tools (Stripe, Slack, CRM, Email) using frameworks like MCP (Model Context Protocol).
• The Orchestration Layer: As noted by Scott Belsky, the "orchestration layer" is the new interface. You must build systems that coordinate agents, handle retries, and verify outputs.

Phase 4: Scaling and Monetization (Steps 19–30)

• High-Touch Onboarding: Use onboarding to gather data that creates a "moat" (user preferences and long-term memory).
• Outcome Pricing: Move away from "per-seat" pricing. Charge per task completed. This is the future of AI-driven software.
• Compound Value: As the product matures, add adjacent workflows, increase pricing, and turn power users into public case studies.
• Final Goal: Become the "default execution layer" for your chosen sub-niche.

2. Key Arguments and Perspectives

• SaaS is Evolving, Not Dying: Eisenberg argues that while traditional per-seat SaaS is struggling, the "Agentic" era offers a massive opportunity for smaller, cash-flowing startups.
• The "Conductor" Era: The most valuable software companies will be those that own the "orchestration layer"—the ability to coordinate multiple AI agents to achieve a business outcome.
• Service-to-Software Transition: The most successful startups will start as service businesses (human-led) and gradually replace human labor with AI agents as the workflows are perfected.

3. Notable Quotes

• "The orchestration layer is the new interface layer... The ultimate layer to own is where coordination takes place." — Attributed to Scott Belsky.
• "If coordination becomes the choke point, whoever owns that layer owns the flow."
• "You’re not just going to want to create a product... you’re going to want to create media."

4. Synthesis and Conclusion

The future of SaaS lies in specialization and automation. By targeting a narrow sub-niche, mapping the exact workflow where money changes hands, and building an AI-orchestrated system to execute that workflow, founders can build highly profitable, cash-flowing businesses. The competitive advantage is no longer just the code; it is the media-driven audience and the proprietary data/memory stored within the agentic workflow. Founders should focus on becoming the "default execution layer" for their niche, moving from manual service to automated, outcome-based software.