Usage-based pricing has moved from a cloud-billing curiosity to a default option in modern SaaS. The shift is partly driven by customer demand (no one wants to pay for unused seats), partly by investor preference (usage pricing produces higher net revenue retention), and partly by improved tooling that makes accurate metering feasible for almost any product.
What it looks like in SaaS specifically
- Per-API-call pricing — Twilio, Stripe, OpenAI. Each call equals a billable unit.
- Per-event pricing — Segment, Amplitude. Customer pays per event tracked or per active user.
- Per-credit pricing — Snowflake, Databricks. Compute time consumed maps to credits.
- Per-seat-active pricing — Some collaboration tools charge only for seats that actually log in during the period.
Why SaaS companies adopt it
Three financial benefits typically follow. First, net revenue retention rises because successful customers grow into more spend without renegotiation. Second, acquisition gets easier because the entry price is lower — start small and grow. Third, alignment improves; the vendor is incentivized to drive customer usage, which is the same thing as driving customer success.
The operational complexity
- Metering pipeline. Real-time or near-real-time accurate event capture. Engineering investment is real.
- Forecasting. Revenue prediction is harder because customer usage varies. Most usage-priced SaaS companies invest in usage analytics as much as in sales forecasting.
- Sales compensation. Whom do you pay — the rep who signed the deal, or the success manager who drove usage? Most modern comp plans split between the two.
- Customer health. Declining usage is a churn signal even when the customer has not churned on paper. Best-in-class teams treat it as such.
See SaaS usage-based pricing for an alternate framing and usage-based pricing model for design specifics.