SaaS Industry Structure in the AI Era
Per-seat collapse, usage-based transition, data moat formation โ how AI is restructuring the SaaS industry itself.
The SaaS industry faces a fundamental revenue model inflection point driven by AI agent proliferation. As per-seat models weaken and usage/outcome-based pricing emerges, the bifurcation between data-moat platforms and feature-selling tools is accelerating.
Market Size
Global SaaS market: ~$320B in 2025 โ projected $820B by 2030 (CAGR ~20%)
Key Trends
Per-seat โ Usage/Outcome Shift
Share of companies using outcome-based pricing jumped from 2% to 18% in 2Q25. 37% plan to change their pricing model within a year.
AI Agent Execution Layer Demand
As AI agents scale, structural demand grows for platforms (ServiceNow, Salesforce) that handle execution on legacy systems.
Data Accumulation Platform Bifurcation
Platforms where customer data accumulates become AI substrates; feature-only tools get eaten by AI startups. Gartner projects 35% of point-product SaaS to be replaced by AI by 2030.
Stock-Earnings Divergence
During the 2025 SaaS stock crash, RPO, NRR, and ARR were actually rising. Market fear outran actual fundamentals.
Key Players
The SaaS market divides into horizontal and vertical SaaS, with each requiring a different survival strategy in the AI era.
Segments
Horizontal SaaS
~60%CRM, HRM, ERP, collaboration tools applied across all industries. Feature commoditization by AI threatens companies without data moats.
Vertical SaaS
~40%Software specialized for specific industries (healthcare, manufacturing, legal). Hard to replace due to regulatory and data specificity, but AI-native competitors are emerging.
Value Chain
Snowflake, MongoDB, AWS S3
AI workload growth driving consumption surge
Snowflake, Databricks, dbt
Growing demand for real-time processing
Databricks, AWS SageMaker, Azure ML
Expanding LLM fine-tuning demand
Salesforce, ServiceNow, HubSpot
AI interfaces users interact with directly
Datadog, Dynatrace, New Relic
AI workload monitoring demand growing automatically
Interest rates, AI infrastructure investment, and enterprise IT budgets are the core macro variables for the SaaS industry.
Rate Cut Expectations
High-growth SaaS multiple re-rating โ stock re-rating
AI Infrastructure Investment Growth
Cloud/SaaS consumption growth โ usage-based revenue growth
Per-seat Model Erosion
AI agents replacing employees โ seat count reduction
Enterprise IT Budget Freeze
Economic uncertainty โ software spending optimization โ NRR pressure
AI Startup Price Competition
ARPU pressure on feature-focused SaaS
Data Sovereignty Regulations
Data localization requirements โ increased costs, but strengthens local platforms
SaaS won't die. But it will bifurcate. Companies with data moats and usage-based transitions will benefit from AI; feature-only companies will face structural pressure.
Opportunities
- Monopoly positioning of AI agent execution layer platforms (ServiceNow, Salesforce)
- Direct AI workload growth benefit for usage-based companies (Snowflake)
- Observability demand (Datadog) โ scales proportionally with AI infrastructure
- AI app data layer (MongoDB Atlas) โ becoming a developer default
- ARPU re-acceleration for companies completing outcome-based transitions
Risks
- Accelerating structural revenue erosion for per-seat model companies
- Microsoft vertically integrating M365 + Copilot + Dynamics threatens multiple SaaS positions
- AI startup feature commoditization weakening pricing power of incumbent SaaS
- IT spending cut first in recessions, potentially damaging short-term growth
Verdict
Verdict: "Avoid all SaaS" is wrong. "SaaS is fine" is also wrong. Which SaaS is everything. Filter by three criteria: revenue model transition, data accumulation structure, and actual quarterly results.