SaaS Pricing Models: What Works and When

Every SaaS founder hits the same wall: you’ve built something people want, and now you need to figure out how to actually charge for it without driving those people away.

The pricing model you choose isn’t just a revenue decision. It’s a product decision, a retention decision, and a growth decision rolled into one. Get it wrong and you’ll spend the next six months rewriting your billing system, retraining your customers, and watching churn compound.

This isn’t a theoretical exercise. If you’re a technical founder or engineering lead shipping your own SaaS product, your pricing model determines what you need to build on the back end — usage metering, seat tracking, overage calculations, proration logic — and how your customers experience your product on the front end.

Let’s break down the pricing models that actually work, who should use each one, and how to implement them without becoming a part-time billing engineer.

The Five SaaS Pricing Models That Actually Work

1. Per-Seat Subscription (The Classic B2B SaaS Pricing Model)

  • Your product’s value is tied to the number of people using it
  • Your customers are teams or organizations with 2+ users
  • Your marginal cost per additional seat is near-zero
  • Customers work around seat limits (shared logins, which kills your usage data and compliance)
  • Heavy users and light users pay the same, creating dissatisfaction on both sides
  • You’re building an AI tool where the value isn’t “more people” but “more outputs”

2. Tiered Subscription (Feature-Gated)

  • Your product has clear, defensible feature tiers
  • You can articulate real differences between tiers (not just “more” of everything)
  • Your customers self-select based on their needs without heavy sales involvement
  • You gate features that customers expect to be standard (this is how you get 1-star reviews)
  • The gap between tiers is too large or too small
  • Your implementation makes it painful to upgrade mid-cycle

3. Usage-Based Pricing (TheSaaS Consumption Model)

  • Your marginal cost per unit is measurable and relatively predictable
  • Customer usage varies significantly (one-size-fits-all would alienate either light or heavy users)
  • You’re building infrastructure, developer tools, or AI products where consumption = value
  • Your customers need predictable budgeting (finance teams hate variable costs)
  • You don’t have robust usage metering (bill shock kills trust fast)
  • Your unit economics are thin and volume-driven (a 10% pricing error at scale is catastrophic)

[Internal link: See our deep dive on usage-based billing implementation for the technical walkthrough.]

4. Hybrid Pricing (Base Subscription + Usage Overage)

  • Most SaaS products, especially AI-powered ones
  • You want predictable baseline revenue but need to capture value from power users
  • Your customers want budget certainty but expect to pay more when they scale
  • The included allowance is too low (customers hit overage immediately and churn) or too high (you leave money on the table)
  • Your implementation can’t handle both subscription lifecycle and usage metering simultaneously

[Internal link: Read more about consumption-based pricing models for SaaS for a detailed breakdown of the consumption model.]

5. Per-Value / Outcome-Aligned Pricing

  • Your product directly enables measurable customer revenue or cost savings
  • You can integrate with or observe the customer’s revenue/success metrics
  • You’re comfortable with variable revenue in exchange for lower customer acquisition friction
  • You can’t reliably track or attribute the outcome you’re pricing on
  • Your customers are unwilling to share their revenue data with you
  • The outcome metric has high variability outside your control

SaaS Business Model Metrics: What to Track

No matter which pricing model you choose, you need to track these metrics to know if it’s working:

These SaaS business model metrics determine whether your pricing model is actually serving your growth or just creating the illusion of revenue.

The Hidden Cost: Building SaaS Billing Infrastructure

Here’s what most SaaS pricing model comparisons don’t cover: the implementation cost of each pricing model.

Per-seat pricing requires user management and seat counting. Tiered pricing adds feature flags and proration logic. Usage-based pricing demands real-time metering, aggregation, and variable invoicing. Hybrid pricing combines all of the above. Outcome-aligned pricing requires revenue tracking and attribution.

Most founders build these systems by stitching together Stripe for SaaS billing, Auth0 for authentication, custom code for metering, and a spreadsheet for reporting. That’s 4-6 integrations, each with its own webhooks, error handling, and reconciliation logic. By the time you’ve built it, debugged it, and made it production-ready, the process often consumes significant engineering time and budget.

The alternative is a unified platform that handles authentication, billing, user management, usage tracking, custom subscription logic, and reporting in one integration. That’s 1 integration instead of 6, with built-in support for all five pricing models we’ve discussed.

You build. Someone else should monetize.

FAQ

What is the most common SaaS pricing model?

Per-seat subscription remains the most common B2B SaaS pricing model . However, usage-based pricing is growing rapidly, and hybrid models (subscription + usage) are increasingly adopted by companies looking to move beyond pure per-seat pricing.

It depends on your cost structure. If your marginal cost per user is near-zero (most software), per-seat works fine. If your marginal cost varies significantly with usage — especially for AI products where inference costs scale with token consumption — usage-based or hybrid pricing is more sustainable. The saas consumption model is particularly well-suited to AI products because usage directly correlates with both cost and value.

Start with your unit economics, not your competitors. Calculate your customer acquisition cost, your cost to deliver the product, and your target LTV:CAC ratio. Work backward from those numbers to determine your minimum viable price, then test price points with your earliest customers. Competitor pricing is a reference point, not a strategy.

A consumption model in SaaS is a pricing approach where customers pay based on how much of the service they consume, measured in units like API calls, compute hours, messages processed, or storage used. Unlike subscription pricing where the customer pays a flat fee regardless of usage, the saas consumption model ties revenue directly to customer activity. This model is gaining traction in AI SaaS, developer tools, and cloud infrastructure — sectors where usage varies dramatically between customers and correlates strongly with value delivered.

Proration splits the billing cycle proportionally. If a customer on a $30/month plan upgrades to $99/month on day 15 of a 30-day cycle, they’ve already paid for 15 days at $30 ($15 used, $15 credit), so they owe ($99 – $30) × (15/30) = $34.50 prorated. Most modern billing platforms handle this automatically, but custom implementations often get it wrong — off-by-one errors, timezone mismatches, and currency rounding are common failure points.

It depends on your cost structure. If your marginal cost per user is near-zero (most software), per-seat works fine. If your marginal cost varies significantly with usage — especially for AI products where inference costs scale with token consumption — usage-based or hybrid pricing is more sustainable. The saas consumption model is particularly well-suited to AI products because usage directly correlates with both cost and value.

Start with your unit economics, not your competitors. Calculate your customer acquisition cost, your cost to deliver the product, and your target LTV:CAC ratio. Work backward from those numbers to determine your minimum viable price, then test price points with your earliest customers. Competitor pricing is a reference point, not a strategy.

A consumption model in SaaS is a pricing approach where customers pay based on how much of the service they consume, measured in units like API calls, compute hours, messages processed, or storage used. Unlike subscription pricing where the customer pays a flat fee regardless of usage, the saas consumption model ties revenue directly to customer activity. This model is gaining traction in AI SaaS, developer tools, and cloud infrastructure — sectors where usage varies dramatically between customers and correlates strongly with value delivered.

Proration splits the billing cycle proportionally. If a customer on a $30/month plan upgrades to $99/month on day 15 of a 30-day cycle, they’ve already paid for 15 days at $30 ($15 used, $15 credit), so they owe ($99 – $30) × (15/30) = $34.50 prorated. Most modern billing platforms handle this automatically, but custom implementations often get it wrong — off-by-one errors, timezone mismatches, and currency rounding are common failure points.

Start with your unit economics, not your competitors. Calculate your customer acquisition cost, your cost to deliver the product, and your target LTV:CAC ratio. Work backward from those numbers to determine your minimum viable price, then test price points with your earliest customers. Competitor pricing is a reference point, not a strategy.

A consumption model in SaaS is a pricing approach where customers pay based on how much of the service they consume, measured in units like API calls, compute hours, messages processed, or storage used. Unlike subscription pricing where the customer pays a flat fee regardless of usage, the saas consumption model ties revenue directly to customer activity. This model is gaining traction in AI SaaS, developer tools, and cloud infrastructure — sectors where usage varies dramatically between customers and correlates strongly with value delivered.

Proration splits the billing cycle proportionally. If a customer on a $30/month plan upgrades to $99/month on day 15 of a 30-day cycle, they’ve already paid for 15 days at $30 ($15 used, $15 credit), so they owe ($99 – $30) × (15/30) = $34.50 prorated. Most modern billing platforms handle this automatically, but custom implementations often get it wrong — off-by-one errors, timezone mismatches, and currency rounding are common failure points.

A consumption model in SaaS is a pricing approach where customers pay based on how much of the service they consume, measured in units like API calls, compute hours, messages processed, or storage used. Unlike subscription pricing where the customer pays a flat fee regardless of usage, the saas consumption model ties revenue directly to customer activity. This model is gaining traction in AI SaaS, developer tools, and cloud infrastructure — sectors where usage varies dramatically between customers and correlates strongly with value delivered.

Proration splits the billing cycle proportionally. If a customer on a $30/month plan upgrades to $99/month on day 15 of a 30-day cycle, they’ve already paid for 15 days at $30 ($15 used, $15 credit), so they owe ($99 – $30) × (15/30) = $34.50 prorated. Most modern billing platforms handle this automatically, but custom implementations often get it wrong — off-by-one errors, timezone mismatches, and currency rounding are common failure points.

Proration splits the billing cycle proportionally. If a customer on a $30/month plan upgrades to $99/month on day 15 of a 30-day cycle, they’ve already paid for 15 days at $30 ($15 used, $15 credit), so they owe ($99 – $30) × (15/30) = $34.50 prorated. Most modern billing platforms handle this automatically, but custom implementations often get it wrong — off-by-one errors, timezone mismatches, and currency rounding are common failure points.