How to set pricing metrics for usage-based and AI SaaS

In usage-based SaaS pricingmodels the invoice for a period is simply: quantity consumed × price per unit. Choosing the right “unit” determines perceived fairness, sales velocity, and net revenue retention (NRR). Lago processes up to 15,000 SaaS pricing models — cloud-hosted with enterprise controls (self-hosted available as an alternative).

Quick pricing examples (lead with real breakpoints)

• Per seat/count: 1–10 users = $25/user/mo, 11–50 = $20/user/mo
• Per API call/transaction: $0.002 per API call, volume discounts at 10M+ calls
• Per GB/volume: $0.12 per GB ingested, $0.05 per GB egress
• Per time: $0.05 per minute for real‑time sessions; billed to nearest second

These examples reflect common saas pricing models and show how breakpoints and multipliers shape customer costs and supplier revenue.

The four primary value metrics

• Time — seconds/minutes/hours/days (duration-based)
• Transaction — API requests, jobs, image processes (discrete events)
• Volume — KB/GB/TB, rows scanned, bytes ingested (continuous)
• Count — devices, hosts, seats, records (items managed)

Providers can price on one metric or combine multiple metrics for a hybrid model. For implementation patterns and architectures that combine subscriptions and usage, see What are hybrid pricing models and how do they work? - Lago.

1) Time-Based Metrics

Units: seconds, minutes, hours, days

When to use

• Services where duration equals value (VM runtime, video conferencing).
• Useful for predictable per-session billing (e.g., $0.05/min).

Business outcomes

• Clear correlation between usage and cost improves customer acceptance.
• Enables peak/off-peak multipliers to capture higher willingness to pay.

Implementation notes

• Meter at fine-grain (seconds) to avoid overcharging.
• Use rate multipliers for SLA tiers (standard vs. expedited).

2) Transaction-Based Metrics

Units: API calls, transactions, queries, test runs

When to use

• API-first products, image/AI inference, message delivery where each call has roughly equal marginal value.

Pricing example

• $0.002 per API call with tiered discounts: 0–1M = $0.002, 1M–10M = $0.0015, 10M+ = $0.001.

Business outcomes

• Aligns revenue to consumption spikes (reduces friction for light users).
• Drives pay-as-you-go adoption and can increase conversion from free tiers.

Implementation notes

• Accurate event deduplication and idempotency are required to avoid billing disputes.
• High throughput metering platform is essential to avoid latency in billing pipelines — Lago supports large-scale event ingestion and complex rating rules. See SaaS billing system guidance: SaaS Billing Systems That Handle Complex Pricing Models.

3) Volume-Based Metrics

Units: KB/MB/GB/TB, number of records scanned

When to use

• Storage, data ingestion, egress, monitoring, and analytics where value scales with bytes or records.

Pricing example

• $0.12 per GB ingested; $0.05 per GB egress; first 100 GB free or included.

Business outcomes

• Predictable for data-heavy customers when combined with retention/archival tiers.
• Encourages data hygiene policies (customers optimize what they store/transfer).

Implementation notes

• Plan for retention windows and tiered storage pricing.
• Meter both ingestion and long-term storage separately to reflect operational cost.

4) Count-Based Metrics

Units: connected devices, hosts, seats, records

When to use

• IoT, monitoring, security where value is per‑item management or protection.

Pricing example

• $3 per device/month for first 100 devices, $1.50 thereafter.

Business outcomes

• Simple to explain and sell; easy to forecast revenue by customer headcount or device fleet.
• Dynamic counts require continual reconciliation to avoid over/under billing.

Implementation notes

• Support day‑level reconciliation and prorations when counts change within billing cycles.

Rate Multipliers (secondary modifiers)

Common multipliers:

• Time of day (peak vs. off‑peak)
• SLA/performance (standard vs. expedited)
• Outcome (success vs. failed verification)
• Environment (production vs. test)

Example: $0.10 per verification success, $0.01 for failures.

Business effect: multipliers capture willingness to pay for scarce capacity or higher certainty and increase revenue per high-value use cases.

Multi-Dimensional Pricing

Two common patterns:

1. N‑metered variables (transactions + volume + time)
2. Metered + non‑metered (fixed base fee + metered usage)

Example: $500 base platform fee + $0.001 per API call + $0.12 per GB.

Why multi-dimensional

• Captures multiple value drivers, reduces churn by providing a baseline commitment, and improves monetization of high-usage customers.

Operational note: multi-dimensional pricing increases billing complexity — a billing platform must support composite rating, discounts, and consolidated invoices to keep invoices intelligible for customers and to speed time‑to‑cash. For practical models and AI use cases, review 6 Proven Pricing Models for AI SaaS (and How to Build Them).

How to pick the right metric (checklist)

1. Align metric to primary customer value (what they measure internally).
2. Ensure metering feasibility (can the platform capture it reliably and at scale?).
3. Evaluate predictability for customers (does metric produce wildly variable bills?).
4. Consider sales motion and comparables (what buyers expect in the category).
5. Account for operational cost (metering, reconciliation, dispute handling).

SaaS founders should prototype with real usage data and iterate pricing breakpoints before a broad roll‑out.

Implementation implications and expected outcomes

• Faster time-to-cash: usage billing with automated invoicing reduces manual reconciliations.
• Higher NRR: aligning price with value keeps heavy users from churning.
• Fewer billing disputes: precise, auditable metering and transparent invoices reduce support load.

Lago’s platform handles high-throughput metering, complex rating (multi-dimensional and multipliers), and automatic invoice generation to reduce billing errors and accelerate financial close. For more on pay-as-you-go architectures and credit models, see Understanding the Pay-as-You-Go Pricing Model for ... - Lago.

External guides on saas pricing models and strategy provide broader market context and best practices.

Call to action

• Evaluate which metric maps to your customer value and run a pricing experiment with a billing engine capable of high-throughput metering. Learn how Lago implements these patterns and accelerates monetization on the platform: Lago or explore implementation guides and examples on the Lago blog: SaaS Billing Systems That Handle Complex Pricing Models.