What is a SaaS pricing model?
A SaaS pricing model is the structure that decides how a customer's bill is calculated - not what the number is. Those are two separate decisions, and conflating them is why pricing conversations go in circles. The model is whether you charge per seat, per unit of usage, per package of features or some combination. The price is what you put on it afterwards.
The model matters more than the number, because it determines how revenue behaves as customers succeed. Get it right and accounts grow on their own as they get more value. Get it wrong and every expansion needs a renegotiation, or your largest customers pay the same as your smallest.
Underneath every model is a value metric: the unit you charge for. Choosing it is the actual decision. Everything else - tiers, discounts, annual terms - is packaging on top.
What are the main SaaS pricing models?
Six structures cover almost every SaaS business. Most mature companies end up with some version of the last one.
| Model | How it works | Best for |
|---|---|---|
| Flat-rate | One price, one product, every customer pays the same. | Single-persona products with little variation in usage. |
| Tiered | Two to four packages at rising prices, each with more features or higher limits. | Most B2B SaaS. It is the default for a reason - it segments willingness to pay without complex metering. |
| Per-seat | Price scales with the number of users on the account. | Collaboration tools where value grows with headcount. |
| Usage-based | Customers pay for what they consume: API calls, events, gigabytes, transactions. | Infrastructure and API products where cost of delivery tracks usage. |
| Freemium | A permanently free tier, with paid plans above it. | Products with low marginal cost, fast time-to-value and natural virality. |
| Hybrid | A tiered base plus usage-based overages, or seats plus consumption. | Businesses that want predictable revenue and expansion that follows real value. |
Flat-rate pricing
One price, one product, every customer pays the same. Single-persona products with little variation in usage.
Watch for: Leaves money on the table at the top and prices out the bottom. Almost nobody stays here past early traction.
Tiered pricing
Two to four packages at rising prices, each with more features or higher limits. Most B2B SaaS. It is the default for a reason - it segments willingness to pay without complex metering.
Watch for: Tier sprawl. Every deal that needs a custom tier is a sign the packaging no longer matches the market.
Per-seat pricing
Price scales with the number of users on the account. Collaboration tools where value grows with headcount.
Watch for: Customers ration seats to control spend, which suppresses adoption - the opposite of what you want.
Usage-based pricing
Customers pay for what they consume: API calls, events, gigabytes, transactions. Infrastructure and API products where cost of delivery tracks usage.
Watch for: Revenue becomes harder to forecast, and customers get bills they did not expect.
Freemium pricing
A permanently free tier, with paid plans above it. Products with low marginal cost, fast time-to-value and natural virality.
Watch for: You are funding the free tier from paid revenue. If conversion is weak, it is a cost centre with a growth story attached.
Hybrid pricing
A tiered base plus usage-based overages, or seats plus consumption. Businesses that want predictable revenue and expansion that follows real value.
Watch for: Complexity. Every extra dimension is one more thing a prospect has to model before they can buy.
What is tiered pricing, and how many tiers should you have?
Tiered pricing packages features and limits into two to four plans at rising prices. It is the default in B2B SaaS because it does two jobs at once: it segments customers by willingness to pay, and it gives every account a visible next step.
Three tiers is the common shape, and there is a reason it persists. One plan gives buyers nothing to compare against. Two forces a binary yes-or-no. Three creates a middle option that most buyers choose, with a cheaper plan making it look reasonable and a premium plan making it look sensible.
The failure mode is tier sprawl. When sales keeps needing a custom package to close, that is not a discounting problem - it is the packaging telling you it no longer matches how the market segments.
What is usage-based pricing?
Usage-based pricing charges for what a customer consumes: API calls, events processed, messages sent, gigabytes stored. It has spread well beyond infrastructure because it aligns the bill with delivered value - customers who get more, pay more, without anyone renegotiating.
The trade-offs are real. Revenue becomes harder to forecast, because next month depends on customer behaviour rather than a contract. Customers dislike unpredictable bills, which is why most usage-based businesses add a committed minimum. And it complicates your recurring revenue definition: usage overages are not recurring in the way a subscription is, so folding them into MRR overstates how predictable your revenue actually is.
Most businesses that adopt it end up hybrid - a committed base that behaves like a subscription, plus consumption above it that behaves like expansion.
Is freemium a pricing model or a marketing strategy?
Both, and treating it as only the first is how it goes wrong. A free tier is a distribution channel you pay for in infrastructure and support, justified only if it produces paying customers more cheaply than buying them.
Freemium works when marginal cost per free user is near zero, the product delivers value without hand-holding, and free users bring in other users through sharing or network effects. Miss any of those and you have a cost centre with a growth story attached.
The honest comparison is against a time-limited free trial, which creates urgency that a permanent free plan never does. We cover that trade-off in detail in freemium vs free trial, and the number that settles the argument is trial-to-paid conversion rate.
What pricing strategies sit on top of the model?
Once the structure is set, the strategy decides what number goes on it.
Value-based pricing
Prices from the economic value the customer receives - what the product saves or earns them - and captures a share of it. Hardest to implement, consistently produces the highest prices, and the only strategy that scales with how good your product gets.
Cost-plus pricing
Takes cost of delivery and adds a margin. Defensible and almost always too cheap in software, where gross margins run 75-85% and cost bears almost no relationship to value.
Competitor-based pricing
Anchors to what similar products charge. Fast, and it quietly cedes your pricing decision to a competitor who may have completely different costs, funding and strategy.
Penetration pricing
Launches deliberately below market to win share, then raises prices later. It can work in crowded categories with low switching costs, but it selects for the most price-sensitive customers - exactly the ones who churn when prices rise - and it anchors what buyers think your product is worth.
How do you choose a pricing model?
Work from the value metric outward. Three questions get you most of the way:
- What grows as a customer succeeds? More people, more volume, or more capability. That answer is your value metric, and it points at seats, usage or tiers respectively.
- Can the customer predict their bill? If not, you will need a committed minimum or a cap. Finance teams reject unpredictable software spend regardless of how fair the metering is.
- Does the model punish adoption? Per-seat pricing makes customers ration seats. If your model gives someone a reason to use the product less, it is working against you.
Model the change before committing to it. Our SaaS pricing calculator shows how plan price, churn and payment fees interact across your MRR and ARR, and how to price your product walks through the exercise end to end.
How does your pricing model change your metrics?
This is the part most pricing guides skip, and it is the part that bites. The model you choose decides which metrics are meaningful and which quietly mislead you.
- Per-seat pricing manufactures expansion. Accounts grow without a sales conversation, which shows up as expansion MRR and pushes net revenue retention above 100%. It also means headcount changes at your customers drive your revenue.
- Usage-based pricing destabilises MRR. A metric built on the assumption of a predictable monthly subscription gets noisy when half the bill is consumption. Separate committed revenue from overage, or your MRR growth rate measures customer behaviour rather than business growth.
- Freemium distorts every per-customer number. Free users inflate customer counts, which drags ARPU down and makes churn rate look better or worse depending on whether free accounts are counted. Decide explicitly, then be consistent.
- Annual billing changes your cash, not your MRR. A yearly plan is normalised to a monthly figure, so MRR barely moves while cash collected jumps. It also collapses CAC payback, often to zero.
- Tier structure sets your ceiling. If your top tier is priced for your median customer, you cap ACV no matter how well sales performs.
Whatever model you land on, the test is the same: does revenue rise when customers get more value, without anyone having to negotiate? The full set of definitions is in the SaaS metrics reference.
Frequently asked questions
What is the most common SaaS pricing model?
Tiered pricing, usually three packages, is the most widely used model in B2B SaaS. It segments customers by willingness to pay without requiring usage metering, and it gives sales a natural upgrade path. Most businesses that start flat-rate or per-seat end up tiered or hybrid within a couple of years.
What is the difference between tiered and usage-based pricing?
Tiered pricing charges a fixed amount for a package of features or limits, so the customer knows the bill in advance. Usage-based pricing charges for what is consumed, so the bill varies month to month. Tiered gives you predictable recurring revenue; usage-based ties revenue to the value delivered but makes forecasting harder.
How do I choose a SaaS pricing model?
Start from your value metric - the thing that grows as a customer gets more out of the product. If that is people, price per seat. If it is volume, price on usage. If it is capability, price in tiers. The model should make a customer’s bill rise as their success rises, without requiring a sales conversation every time.
Is freemium worth it?
Only when three things are true: marginal cost per free user is near zero, the product delivers value without onboarding, and free users create distribution through sharing or network effects. If any of those is missing, a time-limited free trial almost always outperforms freemium, because it creates urgency rather than a permanent free habit.
What is penetration pricing?
Penetration pricing means launching deliberately below market rate to win share quickly, then raising prices later. It can work in crowded categories where switching costs are low, but it attracts the most price-sensitive customers, and those customers churn hardest when prices rise. The discount also anchors what your product is worth.
What is value-based pricing in SaaS?
Value-based pricing sets the price from the economic value the customer receives rather than from your costs or competitors’ rates. In practice it means quantifying what the product saves or earns a customer and capturing a share of it. It is the hardest model to implement and consistently produces the highest prices.
How often should I change my pricing?
Most SaaS businesses revisit pricing far too rarely. An annual review is reasonable, with changes applied to new customers first and existing customers grandfathered or migrated deliberately. What matters more than frequency is measuring the effect: a price change that lifts ARPU while raising churn may leave you worse off.
Does my pricing model affect my SaaS metrics?
Substantially. Per-seat pricing produces expansion MRR as teams grow. Usage-based pricing makes MRR volatile and complicates the definition of recurring revenue. Freemium depresses ARPU and inflates customer counts, which distorts any per-customer metric. The model you choose determines which metrics are meaningful and which mislead.
