SaaS pricing models at a glance
SaaS pricing models decide how a customer’s bill is calculated: a fixed amount per plan, an amount per user, an amount that changes with quantity, or an amount based on actual usage. The model you choose shapes how revenue grows with each customer, how predictable your MRR is, how churn shows up, and what a buyer will see when they review the business.
Stripe’s billing documentation groups recurring pricing into four families (Stripe: Recurring pricing models):
- Flat rate: the customer picks a tier (for example Basic, Starter, or Enterprise) and pays a fixed price. Typical fit: simple tools, small teams, solo founders.
- Per-seat: each pricing unit represents one user. Typical fit: collaboration and team software.
- Tiered: the unit price changes with quantity (volume) or usage (graduated). Typical fit: products with a clear unit that scales.
- Usage-based: customers pay based on usage, including fixed fee plus overage, pay as you go, and credit burndown. Typical fit: APIs, infrastructure, and AI products.
Most real products combine these. A common pattern is a flat or per-seat base plan with a usage-based component for heavy users. This guide explains each model, shows the arithmetic, and covers how each one affects the metrics founders and buyers rely on.
Flat-rate pricing
Each plan has one price. A $29/month plan costs $29 no matter how many people use it or how much they do.
Strengths: easy to explain, easy to forecast, and MRR is straightforward: number of customers on each plan times its price.
Weaknesses: revenue per customer does not grow as the customer gets more value, so expansion revenue comes only from plan upgrades. Heavy users can be unprofitable if your costs scale with usage.
Metric effect: churn is mostly logo churn. Net revenue retention tends to be limited unless upgrades are common. See net revenue retention.
Per-seat pricing
The customer pays per user. Stripe describes it as each pricing unit representing one user, as when a business buys a license for each employee (Stripe: Recurring pricing models).
Example: $12 per seat per month. A 10-person team pays $120; when it grows to 25, it pays $300.
Strengths: revenue grows automatically as customers hire, which creates expansion MRR without a sales conversation. Buyers understand it immediately.
Weaknesses: customers may share logins or limit seats to save money. If the product’s value does not rise with headcount (an automation tool that replaces work, for example), seats can be the wrong unit. Layoffs at customers show up as contraction.
Metric effect: expansion and contraction become large parts of your MRR movement, so track them separately from new and churned MRR.
Tiered pricing: volume vs graduated
Tiered pricing changes the unit price as quantity grows. Stripe distinguishes two methods, using an example with tiers of $7 per unit for units 1–5, $6.50 for 6–10, and $6 for 11+ (Stripe: Tiered pricing):
Volume-based: the whole quantity is billed at the rate of the tier it ends in.
- 5 units: 5 × $7 = $35
- 6 units: 6 × $6.50 = $39
- 20 units: 20 × $6 = $120
Graduated: each tier’s units are billed at that tier’s rate, then summed.
- 5 units: $35
- 6 units: $35 + $6.50 = $41.50
- 20 units: $35 + $32.50 + $60 = $127.50
Stripe notes that with volume pricing, because the tier price applies to the entire quantity, “the total might decrease” when a customer crosses into a cheaper tier. That can create odd incentives, such as a customer adding a unit to lower the bill. Graduated pricing avoids that cliff and is easier to defend, but the bill is harder to predict.
Tiers can also include flat fees. Stripe’s example combines unit prices with a flat amount per tier, so a 12-unit volume customer pays 12 × $3 + $30 = $66, while a graduated customer pays $111 under the same tier table (Stripe: Tiered pricing).
Usage-based pricing
Customers pay for what they consume: API calls, compute minutes, messages, tokens, or storage. Stripe lists three patterns: fixed fee and overage, pay as you go, and credit burndown (Stripe: Recurring pricing models).
Strengths: price tracks value and cost closely. It suits products where usage varies widely between customers, and it lowers the barrier for small customers to start.
Weaknesses: revenue is less predictable. A customer’s bill can drop sharply in a quiet month without them churning. Customers may also fear unpredictable bills, which can slow adoption unless you offer caps, alerts, or prepaid credits.
Metric effect: this is where MRR definitions get tricky. Stripe’s standard MRR calculation “exclude[s] … any metered (usage based) products,” and Stripe lists usage-based MRR as a separate preview feature (Stripe Billing analytics). If much of your revenue is usage-based, a dashboard MRR figure may understate total recurring revenue, while trailing revenue will capture it but fluctuate. Be explicit about which number you are quoting.
Freemium, trials, and one-time pricing
These are not pricing models in the billing sense, but they shape how customers enter a pricing model.
- Freemium: a permanently free tier with paid upgrades. Free users do not count in MRR; Stripe excludes free plans from its MRR and active subscribers (Stripe Billing analytics).
- Free trials: time-limited access before payment. Trial users are also excluded from Stripe’s MRR until they convert.
- Lifetime deals and one-time licenses: cash now, no recurring revenue. They can inflate trailing revenue in a launch month while adding nothing to MRR. A buyer will separate them out. See MRR vs ARR.
Monthly vs annual billing
Billing interval is a separate decision from the pricing model, but it changes the numbers.
Annual plans reduce the number of payment transactions, which matters when fees include a fixed component. On Stripe’s standard US pricing as of October 2026, a domestic card payment costs 2.9% + 30¢ (Stripe US pricing). Twelve monthly $10 payments incur 12 fixed 30¢ charges; one $120 annual payment incurs one. The Stripe fees guide works through this in detail.
Annual plans also change how revenue looks:
- MRR normalizes an annual plan to a monthly value. Stripe’s example: 50 subscribers on a $600/year plan contribute $50 each per month to MRR.
- Trailing 30-day revenue, RevenueBug’s headline metric, records the full annual payment in the month it is collected. A month with many annual renewals can look much larger than a typical month. The methodology explains why this metric is not necessarily MRR.
How to choose a SaaS pricing model
Work through these questions in order:
- What unit best tracks the value a customer gets? Users, projects, contacts, transactions, API calls, or outcomes. If you cannot name one, flat-rate tiers are usually the safest start.
- Do your costs scale with usage? If heavy users cost you real money (AI inference, compute, messaging), a usage component protects margins.
- How predictable do customers need their bill to be? Finance teams at larger customers often prefer fixed commitments; developers testing an API may prefer pay as you go.
- How will customers grow? If customers grow in headcount, per-seat captures expansion. If they grow in volume, tiers or usage do.
- Can your billing system support it cleanly? Complex models increase invoicing errors, support tickets, and disputes.
- What will a buyer or investor see? Predictable recurring revenue is generally easier to value than volatile usage or one-time sales. Use the MRR and ARR calculator and SaaS valuation calculator to see how different revenue mixes change the picture.
How to test pricing changes safely
- Change one variable at a time (price, unit, or packaging) so you can attribute the result.
- Measure over a full billing cycle or longer. Trial conversion may rise while 90-day retention falls.
- Grandfather existing customers or give notice. Unannounced increases cause churn and disputes. Platform rules also matter: Apple notes that price increases for auto-renewable subscriptions “may require user consent” (Apple: Auto-renewable subscriptions).
- Watch refund and dispute rates after a change. A spike suggests customers were surprised. See SaaS chargebacks and refunds.
- Track cohort retention, not just topline MRR, with the churn and retention calculator.
Common mistakes
Pricing on cost instead of value. Cost sets a floor, not a price.
Too many tiers. Customers stall when they cannot tell which plan fits.
Choosing a unit customers can game. Shared logins undermine per-seat pricing; batching can undermine per-request pricing.
Calling usage or one-time revenue MRR. Label metered revenue, lifetime deals, and setup fees separately.
Never revisiting price. Early prices often stay in place long after the product has improved. Review pricing at least annually.
Frequently asked questions
What are the main SaaS pricing models?
Flat rate, per-seat, tiered (volume or graduated), and usage-based. Stripe’s billing documentation lists these four, with usage-based covering fixed fee plus overage, pay as you go, and credit burndown. Many products use a hybrid.
What is the difference between volume and graduated tiered pricing?
Volume pricing bills the entire quantity at the rate of the tier it ends in. Graduated pricing bills each tier’s units at that tier’s rate and adds them up. In Stripe’s example, 6 units cost $39 under volume pricing and $41.50 under graduated pricing.
Is usage-based revenue counted in MRR?
It depends on the system. Stripe’s standard MRR excludes metered (usage-based) products and offers usage-based MRR as a separate preview feature. Always state whether usage revenue is included.
Which SaaS pricing model is best for a micro-SaaS?
Simple flat-rate tiers are usually easiest to build, explain, and support for a small team. Add per-seat or usage components only when a clear value unit or cost driver justifies the complexity. See what is micro-SaaS.
Should I offer annual plans?
Annual plans can reduce payment-fee overhead and improve cash flow, but they make monthly cash revenue lumpier. Report normalized MRR alongside trailing revenue so both views are visible.
How does pricing affect a SaaS valuation?
Buyers generally give more credit to predictable, recurring revenue with healthy retention than to volatile or one-time revenue. The pricing model influences all three. See micro-SaaS valuation.