Why 96.4% of SaaS avoid usage-based pricing

9 MIN · 28 MAR 2026 · SAAS RESEARCH LAB

Usage-based pricing has been called the future of SaaS by analysts and VCs for years. Companies like Snowflake, Twilio, and AWS have built massive businesses on consumption models. So why do the overwhelming majority of SaaS products still charge a flat monthly fee?

We analyzed 110 SaaS pricing pages to find out. The answer was striking: usage-based pricing is nearly nonexistent outside of infrastructure and API products.

The pricing model breakdown

Here is exactly how 110 SaaS products structure their pricing.

Flat-rate and per-seat pricing account for 90% of SaaS products. Pure usage-based pricing is a 3.6% minority.

Why flat-rate dominates

1. Revenue predictability for vendors

SaaS companies live by Monthly Recurring Revenue. Flat-rate pricing makes calculation straightforward: customers multiplied by plan price. Usage-based models make forecasting significantly harder because consumption fluctuates month to month.

2. Budget predictability for customers

CFOs want to know exactly what a tool costs next month. A $29/month plan is easy to approve. A usage-based bill that swings from $15 to $150 creates anxiety and buying friction.

3. Simpler sales and marketing

Three or four tiers with clear features is something prospects evaluate in minutes. Usage-based pricing requires calculators, volume estimates, and often a sales conversation — all of which slow down self-serve funnels.

The problem with pure usage-based pricing

• Revenue volatility: 15–25% monthly variance vs 2–5% for flat-rate
• Churn risk: revenue drops immediately when usage drops, not at renewal
• Pricing page complexity: none of the 4 usage-based products had a simple, scannable pricing page
• CTA friction: 3x more likely to require sales contact vs self-serve signup
• Freemium incompatibility: only 1 of 4 usage-based products offered a free tier, vs 54.1% overall

When usage-based pricing actually works

The 4 products using this model share common characteristics:

• Infrastructure products where consumption varies 100x between customers
• API-first products where value scales directly with volume (email, SMS, payments)
• Data processing tools where cost to serve genuinely scales with usage

The common thread: the vendor’s costs scale proportionally with the customer’s usage. When serving a 10x-larger customer costs genuinely 10x more, usage-based pricing aligns incentives correctly.

Hybrid pricing: the emerging middle ground

While pure usage-based pricing is rare (3.6%), hybrid models are emerging among AI-powered SaaS tools. The typical structure:

• A flat monthly fee ($29–99) with a base usage allocation
• Overage charges when the customer exceeds the allocation
• Tied to AI metrics: tokens, API calls, or processing minutes

This gives customers budget predictability while capturing more revenue from heavy users. All 3 hybrid products in our dataset were AI-focused tools launched after 2024.

What this means for your pricing strategy

1. Default to flat-rate tiered pricing unless you have a specific reason not to. It supports PLG, freemium, and predictable revenue.
2. Use per-seat if your product is collaborative. Tools where value grows with team size fit per-seat naturally.
3. Only use pure usage-based pricing if costs genuinely scale with consumption. Infrastructure, APIs, data processing — not typical SaaS apps.
4. Consider hybrid for AI products. Flat base plus usage overage is the emerging pattern.