Why AI Isn’t Killing SaaS Yet

What It Covers

Ramp economist Ara Kharazian analyzes $100 billion in annual business spend data from 50,000 companies to challenge the "SaaS apocalypse" narrative, revealing that seat-based pricing remains dominant at 65–75% of spend, token-based adoption sits below 1%, and traditional SaaS vendors continue growing despite AI competition.

Key Questions Answered

• •SaaS pricing resilience: Seat-based contracts still represent 65–75% of business software spend, with platform fees at roughly 30%. Token-based pricing, despite being offered by companies like HubSpot and Adobe, accounts for less than 0.5% of actual spend on those platforms. Businesses are not meaningfully shifting purchasing behavior toward agentic or consumption-based models yet.
• •Multi-model adoption trajectory: Early AI adopters on Ramp's platform increasingly use multiple models simultaneously rather than committing to one provider. Since early adopters historically predict mainstream behavior, businesses should plan vendor strategies assuming parallel deployment across Anthropic, OpenAI, and emerging cheaper alternatives rather than betting on a single model provider.
• •AI token cost trajectory: High-intensity AI spenders saw token costs increase 13x in one year, reaching roughly 2% of total business spend excluding payroll. This unsustainable growth is pushing cost-conscious firms toward routing platforms like OpenRouter, which directs queries to cheaper or open-source models—a practice likely to spread from heavy users to mainstream businesses.
• •AEO as an emerging category: Answer Engine Optimization—software tracking brand visibility inside AI model responses—is one of the fastest-growing SaaS categories on Ramp's platform. New entrants like Profound are leading this space, not legacy SEO vendors. Businesses investing in AI discoverability should evaluate dedicated AEO tools rather than waiting for existing SEO platforms to catch up.
• •Revenue-headcount decoupling: Early Ramp data shows software companies are growing revenue without proportional headcount increases, suggesting AI-driven productivity gains are beginning to materialize. However, fast-growing AI-adopting firms also tend to have high labor demand, meaning AI adoption correlates with expansion rather than pure workforce reduction at high-performing companies.