Overview
A pricing metric defines what customers pay for - per user, per transaction, per API call
It controls how revenue scales as customers get more value from the product
The strongest metrics align closely with how customers experience value
Most pricing problems come from choosing the wrong metric rather than the wrong price
When the metric is right, the pricing model becomes much easier to design
Pricing conversations in SaaS usually start with the wrong number. Teams spend weeks debating whether to charge $49 or $59. They spend far less time on a more important question - what should they be charging for in the first place?
That second question is about the pricing metric. It defines the unit being monetized, and it determines whether revenue scales as customers use the product or stalls as they grow.
Getting the price wrong by 10% is recoverable. Getting the metric wrong often isn't.
What is a SaaS pricing metric?
A pricing metric - also called a value metric - is the unit a SaaS company charges customers for. It's the “price per ___” inside any pricing model.
Common examples:
- Price per user (Salesforce, Notion)
- Price per transaction (Stripe, PayPal)
- Price per API call (Twilio, OpenAI)
- Price per gigabyte stored (AWS S3, Snowflake)
Even a flat annual enterprise contract has a metric - the account itself.
The metric matters because it controls how revenue grows as the customer gets more value. When usage goes up and the metric goes up with it, the pricing model scales. When they move independently, expansion stalls and discounting creeps in.
Why the metric matters more than the price
Changing a price from $49 to $59 shifts revenue by a fixed percentage. Changing the pricing metric can reshape the entire monetization curve.
Take two versions of the same collaboration tool:
- Charging per seat: revenue grows as teams adopt the product
- Charging per company: revenue flatlines once a customer signs
Same product, same price level. But a very different outcome over time.
This is why the metric is usually the biggest structural lever in a SaaS pricing model - and why fixing pricing often starts here.
Pricing metric vs. pricing model vs. pricing modality
Three terms get used interchangeably in pricing conversations. But in reality, they describe different parts of the same structure:
Pricing metric - what customers pay for
Examples: users, API calls, transactions, storage, active contacts
Pricing modality - how customers pay
Examples: subscription, pay-as-you-go, prepaid credits, hybrid
Pricing model - the combination of the two
Pricing model = pricing metric + pricing modality
The metric defines the unit of value. The modality defines the billing structure around it. Together they form the pricing model.
Most debates inside SaaS companies are about the model. The bigger lever is usually the metric underneath it.
The Value Metric Ladder
Not every metric is equally useful. Some sit right next to the value customers care about, others are several steps removed.
The further a metric is from customer value, the harder it becomes for pricing to scale.
The Value Metric Ladder sorts metrics by how close they sit to customer outcomes:
Infrastructure metrics measure what it costs to run the product. Easy to implement, but customers rarely care about servers - they care about what the product does for them. Pricing conversations drift toward cost rather than value.
Usage metrics get closer. API calls and tokens reflect real product activity. The challenge is connecting usage back to an outcome the customer can justify to their own CFO.
Activity metrics track what customers actually do in the product - transactions processed, orders fulfilled, tickets resolved. These often land well because they map to meaningful work.
Outcome metrics are the closest to value - revenue generated, money saved, leads converted. When pricing is tied to outcomes, vendor and customer move in the same direction. These are powerful, but harder to measure and implement, which is why they remain the rarest of the four.
Higher on the ladder generally means pricing scales more naturally and sells more easily. Lower on the ladder doesn't break the model, but it usually creates more friction.
The 4 parameters of a good pricing metric
Picking a pricing metric means evaluating several candidates against a consistent set of criteria. The Pricing Roadmap by Ulrik Lehrskov-Schmidt defines four parameters:
Operational viability
The metric has to be measurable, definable, and billable. Reliably.
Units like “value delivered” or “business impact” sound appealing but fall apart in practice. Strong pricing metrics are typically product-generated - users, transactions, API calls, storage, compute. If the system can't track it automatically, it will cause billing arguments later.
Works: transactions processed, active users, API calls
Breaks: “outcomes achieved”, “ROI delivered”, anything requiring manual agreement
Value alignment
The metric should increase as customers get more value from the product.
Charging per transaction aligns pricing with economic activity. Charging per active user aligns with real adoption. Charging per API call aligns with product consumption. When the metric moves, value has moved with it.
When the metric sits too far from value, pricing becomes hard to justify in sales and hard to scale in expansion.
Expectation to pay
Even a value-aligned metric fails if customers don't expect to pay for it.
Customers naturally accept paying for:
- Transactions processed
- Users accessing the system
- Messages delivered
Customers resist paying for:
- Internal infrastructure
- Background system operations
- Technical implementation details
The test: does the unit feel like something worth paying for, or does it feel arbitrary? Arbitrary metrics generate resistance in procurement even when the math works out.
Density
Density describes how evenly value is distributed across the units of a metric.
In a high-density metric, every unit carries roughly the same value. Stripe's transaction-based pricing is a textbook example - every dollar processed is worth roughly the same to the customer, which is why the model sells so cleanly.
In a low-density metric, some units are worth far more than others. That's when sales cycles lengthen, discount requests appear, and customers start routing activity around the system.
Density is often the hidden variable that determines whether a pricing model scales or breaks under pressure.
Evaluating a candidate metric
A metric that scores well on all four tends to create pricing that sells cleanly and scales naturally. A metric that scores poorly on two or more is usually the root cause when pricing feels stuck.
Common SaaS pricing metrics
Pricing metrics vary across SaaS products depending on how value is created. The most common patterns fall into four categories.
Seat-based (per user)
Used by: Slack, Salesforce, Notion, Miro
Works when: value grows with team adoption
Breaks when: value doesn't correlate with user count, which leads to seat-rationing, shared logins, and stalled expansion
Usage-based (per unit consumed)
Used by: Twilio, OpenAI, AWS, Snowflake
Works when: consumption maps cleanly to customer value and buyers can predict spend
Breaks when: consumption varies wildly or customers can't forecast cost, which creates procurement resistance and slows expansion
Credit-based (prepaid bundles)
Used by: OpenAI (enterprise), ElevenLabs, Zapier, many AI platforms
Works when: the product has multiple consumption surfaces and customers want billing flexibility
Breaks when: credits expire unused, pricing becomes opaque, or the conversion from credit to outcome is unclear
Transaction-based (per activity)
Used by: Twilio, OpenAI, AWS, Snowflake
Works when: consumption maps cleanly to customer value and buyers can predict spend
Breaks when: consumption varies wildly or customers can't forecast cost, which creates procurement resistance and slows expansion
Usage-based (per unit consumed)
Used by: Stripe, Shopify, PayPal
Works when: each transaction represents clear economic value
Breaks when: transactions vary significantly in value, which pushes customers toward custom deals and complex tiersA common refinement: instead of charging per raw unit, charge per active unit. Slack doesn't charge per user - it charges per active user. That one change improves density by making sure every billed unit reflects real usage.
How successful SaaS companies choose their metric
Looking across category leaders shows a clear pattern: the metric usually reflects how the product creates value rather than how the company incurs cost.
Three patterns show up repeatedly:
Collaboration tools price per user. Value scales with adoption across a team.
Infrastructure and developer platforms price on consumption. Value scales with technical usage.
Commerce and payments platforms price on activity or outcomes. Value scales with economic events the customer already tracks.
The strongest pricing metrics scale with customer value. Vendor cost sits in a different conversation.
How to choose the right pricing metric for your product
Five questions cover most of the work:
What are all the units we could charge for?
List every measurable unit the product generates. Users, transactions, records, events, storage, compute, active accounts. Start wide.
Which of these can we actually track and invoice reliably?
Cut anything the system can't measure automatically. Operational viability removes more pricing candidates than any other factor
Which sit closest to customer value?
Use the Value Metric Ladder. Outcome > activity > usage > infrastructure. Higher on the ladder usually means pricing scales more naturally.
Which feel fair to pay for?
If a buyer in procurement would call the unit arbitrary, it will generate resistance. Test each candidate against the expectation-to-pay bar.
Which has the highest density?
Does each unit carry roughly equal value? If not, is there a refinement that improves it - like Slack's shift from “users” to “active users”?
The best metric usually scores highest across all five questions. It's rarely the one that's easiest to implement.
A word on metric density
Metric density is often the most underappreciated of the four parameters - and the one that separates pricing models that hold up at scale from those that break down under pressure.
High-density metrics sell more cleanly, expand more smoothly, and resist discounting. Low-density metrics create pricing debates that cost more than the pricing itself.
Frequently asked questions
01
What is a pricing metric in SaaS?
A pricing metric - also called a value metric - is the unit a SaaS company charges customers for. It defines the “price per” part of the pricing model, such as price per user, per transaction, per API call, or per gigabyte stored. The pricing metric controls how revenue grows as customers use the product.
02
What is the difference between a pricing metric and a pricing model?
A pricing metric defines what customers pay for. A pricing model describes the overall structure - the metric plus the billing modality (subscription, usage, credits, etc.). In short: pricing model = pricing metric + payment structure.
03
What is a value metric in SaaS?
A value metric is another term for a pricing metric, usually emphasizing that the unit reflects the value customers receive. The closer a metric sits to customer value, the stronger the resulting pricing model tends to be.
04
What are the most common SaaS pricing metrics?
The most common are: seat-based (per user), usage-based (per API call, token, or compute unit), transaction-based (per order or payment), infrastructure-based (per gigabyte or compute hour), and contact-based (per stored record or active customer). Different SaaS categories favor different metrics depending on how value is created.
05
What makes a good SaaS pricing metric?
Four criteria from The Pricing Roadmap: operational viability (measurable and billable), value alignment (scales with customer value), expectation to pay (feels fair to buyers), and density (each unit carries roughly equal value). Metrics that score well across all four produce the most scalable pricing models.
06
Why do SaaS companies use usage-based pricing?
Usage-based pricing lowers the barrier to adoption - customers only pay as they use the product - and scales naturally with consumption. It works well in infrastructure, developer tools, and AI products where usage maps directly to value.
07
What is pricing metric density?
Metric density describes how evenly value is distributed across the units of a pricing metric. High density means each unit carries roughly the same value (like every dollar in a Stripe transaction). Low density means some units are worth far more than others, which creates pricing disputes and slows expansion.
08
What is the difference between usage-based and credit-based pricing?
Usage-based pricing bills customers for actual consumption after the fact (pay-as-you-go). Credit-based pricing has customers buy a pool of credits upfront and draw them down as they use the product. Both reflect underlying usage, but the billing mechanism and customer psychology are different.
09
How do SaaS companies choose the right pricing metric?
By evaluating several candidate metrics against four criteria: can it be measured reliably, does it align with customer value, do customers expect to pay for it, and does each unit carry similar value? The best metric usually scores highest across all four rather than the one that's easiest to implement.
