Is usage-based pricing right for my SaaS business?

A few years ago, shortly after m3ter emerged from stealth, Todd Gardner (SaaS Advisors, and the founder of venture debt firm SaaS Capital) wrote an influential guest blog for us about when usage-based pricing is a good fit.  Tl;dr - a great idea for some SaaS vendors, and a bad idea for others.

When considering usage-based pricing, it's important to evaluate how well this approach aligns with your overall business model, as the right fit can impact scalability and long-term success.

Fast forward to today, and usage-based pricing is no longer the emerging trend it was back then.  It’s become foundational to modern monetization approaches, particularly those associated with AI.  But the core truth Todd identifies still holds - there’s no one-size-fits-all pricing model, and usage-based pricing isn’t always the best option.  

The goal of this refreshed guide is to help you evaluate whether usage-based pricing is right for your business in 2026 - based on today’s realities including the continued growth of AI and economic uncertainty triggered the geopolitical climate. 

Consumption pricing trends in SaaS in 2026

The shift toward usage-based and hybrid pricing models has continued.  PwC and m3ter’s recent survey of 350+ software leaders found that 52% of companies now operate it (ie properly in production, not trialling or experimentation). The usage based pricing report also provides valuable industry data and trends supporting this shift.  

Now the trend towards usage-pricing, often referred to as consumption based pricing, is more established we can start digging into what’s driving it, and the nature of it. This shift includes the adoption of consumption based pricing models, which charge customers based on their actual usage and offer greater flexibility compared to traditional fixed plans.

As you evaluate whether usage-based pricing is right for your business, it’s important to consider how it fits within the broader landscape of saas pricing models.

Buyer expectations have permanently shifted

Customers increasingly expect to pay for what they use. Years of exposure to consumption-based models - from cloud infrastructure to streaming to AI tools - have reset the default expectation in B2B software. Buyers now arrive at the table asking why pricing isn't usage-based, rather than whether it should be.

Customer based pricing models enhance transparency and fairness, making it easier for buyers to understand exactly what they are paying for and increasing overall satisfaction.

Buyers now arrive at the table asking why pricing isn’t usage-based, or why there isn’t a pay as you go option, rather than whether it should be.

Hybrid pricing is the dominant form of usage pricing

Pure usage-based pricing is a minority approach. According to the PwC and m3ter survey data, 72% of B2B companies operating usage-based pricing use a hybrid model - combining a subscription base with usage-based components - while just 28% price on consumption alone. 

The dominance of hybrid reflects a practical reality: customers want the fairness and flexibility of usage-based pricing, but they also want a degree of cost predictability. 

Hybrid models stand in contrast to fixed subscription models, which require customers to commit to a fixed subscription fee upfront, often lacking the flexibility to scale usage as needed.

For vendors, hybrid models offer the best of both worlds too - a reliable revenue base combined with natural expansion as customer usage grows, unlike fixed subscription models that can be rigid and less tailored to individual needs. Hybrid pricing builds upon subscription based pricing, the traditional approach, by adding usage-based elements for greater adaptability.

AI is fundamentally changing software pricing

AI is reshaping not just what software does, but how it gets priced. AI features and services have variable costs - token consumption if you’re built on top of foundational models, inference and compute for the foundational models themselves — so cost to serve varies by usage intensity, and that makes flat-rate pricing risky. 

But the more profound shift is on the value side. AI creates far greater scope to link pricing to outcomes, or meaningful proxies for them: tasks completed, decisions automated, revenue generated. Where traditional SaaS often struggled to find a usage metric that genuinely reflected value, AI-native products frequently have several credible candidates.

Revenue integrity is now a board-level issue

There’s a growing appreciation that moving to usage-based pricing increases revenue integrity risks - ie the risk of revenue leakage due to billing errors, which PwC’s Revenue Integrity team believe is typically 4–7% of ARR.  

This means that revenue growth potential of usage-based pricing is undermined by systems and control failures. To eliminate these, vendors need modern monetisation infrastructure, updated governance, and good preventative controls.

Usage pricing is more resilient in uncertain markets

There is also a macro dimension that has sharpened in 2026. The current geopolitical environment - including the economic uncertainty generated by conflict in the Middle East - could send a chill through the economy. In that context, usage-based pricing offers SaaS vendors a structural advantage: customers facing budget pressure can reduce spend rather than cancel entirely, making logo churn less likely.

Company and product characteristics that lend themselves to usage pricing

Usage-based pricing can be found at all levels of the software stack, from infrastructure through applications. However, there are common characteristics of products with a high concentration of usage pricing. Below is a list of the most common types.

1. High COGS

The classic case in software has been cloud infrastructure providers like AWS, Snowflake, DataDog, MongoDB, Fastly etc.  These companies all deliver lower-in-the-stack cloud functions and incur direct costs when they’re used. These businesses adopted usage-based pricing because ‘fixed revenues but variable costs’ could have put them out of business.  Their customers understand this, so usage pricing also feels fair.

AI services and features similar unit economics - token and inference costs scale with usage.  And similarly, customers understand this, and see usage-based pricing as fair - a natural extension of the utility model. In 2026, this dynamic is arguably the dominant force reshaping SaaS pricing.

2. API-based services

Increasingly software products don't interface with humans - instead they interface with systems via APIs.  For services like these per-seat pricing doesn’t make sense - there is no human user to sit in the seat.  Subscriptions for access is an alternative, but exposes the vendor to the variable cost risk discussed above.  

Usage pricing is often the obvious pricing model, because there are clear usage metrics that are understood by the customer (the API itself) and these align revenues with cost to serve.

3. Unbound upside

Some usage vectors will hit natural limits relatively quickly because there are humans in the loop - lawyers have only so many cases, dentists see only so many patients. But others have no such limits, which means there is wide range of usage intensity.  

AI applications, marketing automation, code testing, and text messaging are strong examples of categories where individual customer usage can grow far faster than the customer's headcount. If your product supports business activity that can scale faster than the customer itself, UBP is worth exploring.

4. Self-provisioning

Most product-led growth (PLG) companies deploy some form of usage pricing for two key reasons.  Firstly, it lowers friction for initial adoption and trial - customers can consume a small amount, across many users, before making a more significant commitment.  Secondly, it removes expansion friction - customers pay more simply by using more. 

5. Products linked to revenue

At the top of the application stack, successful SaaS businesses have identified strong linkages between their product and an increase in their customer's revenue, and deployed usage-pricing to take full advantage. Once the link is established in the customer’s eyes, they feel like the ‘product pays for itself’.  This is the purest form of outcome-based pricing.

Revenue-linked pricing is found in 75% of public SaaS application-layer businesses. Examples include Shopify, HubSpot, Eventbrite, Chargebee, Toast, and Klaviyo. The convergence of SaaS and payments - Shopify being the clearest example — tightens this linkage further and drives organic NRR growth at scale.

5. Obvious proxy for value

Where there is a usage vector that is a good proxy for value derived by the customer, there is a good case for usage pricing, and this is one of the reasons usage pricing is popular for AI products as discussed above.  But it’s not just AI - many older SaaS businesses who had previously defaulted to per-seat pricing have since found better metrics.  One example from Todd's original research: a SaaS vendor focused on frontline worker safety shifted from per-user pricing to the number of times workers used the app for a safety workflow.  This immediately addressed customers' concerns about paying for dormant users.

Per-seat pricing: not always a bad choice

Per-seat pricing gets a bad reputation in UBP conversations, but it isn't universally wrong. It makes sense when the software delivers discrete, individual value - training users, improving personal productivity, or enabling collaboration. Learning systems, HR platforms, and Salesforce automation tools often fit this profile.

The nuance is that many products span both: HubSpot prices its marketing suite by contacts (usage-based) but its sales automation by seat. The right answer is often a hybrid.

What about tiered pricing?

Tiered pricing is a packaging tool, not a pricing philosophy. It can sit on top of either a subscription or a usage-based model. It's useful for customer segmentation and upsell, but it doesn't resolve the fundamental question of whether you should link price to usage. That question has to be answered first.

Expectations when considering usage-based pricing

UBP is not a silver bullet, and transitioning to it involves real operational complexity. A few things to have clear before you start:

Revenue predictability changes shape. UBP revenue is inherently more variable than subscription revenue. You trade some predictability for better alignment - and, typically, better NRR over time. Finance and investors need to understand this shift.

Measurement infrastructure is non-negotiable. You cannot bill for what you can't measure. Reliable usage data capture, metering aligned to contractual terms, and automated billing logic are prerequisites. According to the PwC/m3ter survey, 34% of companies cite improving usage tracking as a top priority over the next 12 months, reflecting the growing revenue capture challenges associated with usage models.

Operational complexity is real. Hybrid and usage pricing is unavoidably more complex than recurring subscriptions, and complexity compounds in larger companies given enterprise sales motions, more intricate billing hierarchies, and greater audit and compliance needs. The risk of errors increases significantly, and the same survey found that 63% of software leaders lack full confidence in their billing data lineage and traceability. 

Getting usage-based billing right requires more than pricing design - it requires the right systems.

The macroeconomic case for UBP is stronger than ever. In a tighter economic environment, buyers favour models where they can control spend without cancelling. UBP gives them that flexibility - and gives you better logo retention as a result.

Implementing usage-based pricing: practical steps and pitfalls

Rolling out a usage-based pricing model is a strategic move that requires careful planning and execution. The first step for SaaS companies is to identify the usage metrics that best reflect the value delivered to customers.. these could include API calls, data storage, or computing power, depending on your product. 

Once you’ve pinpointed the right metrics, design a pricing structure that aligns with your business goals and customer expectations. This might involve tiered pricing, hybrid models that combine subscriptions with usage-based components, or volume-based pricing to reward higher usage volumes.

A robust billing system is essential for accurate usage based billing. Your billing infrastructure must be able to track real-time usage data, automate invoicing, and handle the complexities of hybrid models. This not only ensures accurate billing but also builds trust with customers by minimizing errors and surprises.

Clear communication is another critical factor. Educate your customers about how the usage-based pricing model works, and provide them with real-time access to their usage data. This transparency helps prevent bill shock and enhances customer satisfaction, as customers feel in control of their spending.

Finance teams should be prepared for the operational changes that come with usage-based pricing. This includes adapting revenue recognition processes, updating forecasting models, and managing more variable cash flows. By equipping your finance teams and investing in the right systems, you can avoid common pitfalls and set your SaaS business up for sustainable revenue growth.

Ultimately, successful implementation of a usage based pricing model hinges on aligning your pricing structure with customer value, investing in reliable billing systems, and maintaining open communication. These steps not only enhance customer satisfaction but also drive long-term revenue growth for SaaS companies.

Usage-based pricing: a great idea for some, and a bad idea for others

The intent of the framework above is to give you reference points - not a definitive answer. UBP has been transformative for the right businesses, and genuinely disruptive (in the wrong way) for the wrong ones.

The companies that have succeeded with UBP share a few characteristics: they had measurable usage, products with natural expansion potential, and the operational infrastructure to support consumption-based billing. The ones that struggled typically underestimated the last one.

If you're thinking about UBP - or looking to support a model you've already launched - the right starting point is less about pricing design and more about whether your systems can support what you're designing. Pricing is 50% strategy and 50% operations. Both halves have to be in place.  Talk to m3ter today.

FAQs

Is usage-based pricing only for infrastructure companies?

No. While cloud infrastructure providers like AWS and Snowflake are visible examples, usage-based pricing is now common across the full software stack - from API-based services and AI applications to marketing automation, payments, and PLG SaaS. The key determinant isn't what layer of the stack you're at, but whether you have a measurable usage metric that reflects the value customers receive.

What's the difference between usage-based and hybrid pricing?

Usage-based pricing ties all charges to consumption - you only pay for what you use. Hybrid pricing combines a recurring subscription base with usage-based components on top. According to the 2026 PwC and m3ter survey of 350+ software leaders, 72% of companies operating usage-based pricing use a hybrid model. It's the dominant approach because it balances the fairness and flexibility of usage pricing with the predictability of subscriptions.

How does usage-based pricing affect churn?

It typically reduces logo churn. Customers facing budget pressure can dial usage down rather than cancel entirely - which is significantly better for long-term customer relationships and revenue recovery in future. In uncertain economic environments, this flexibility is a structural advantage of usage-based models over pure subscriptions, where the binary choice is renew or cancel.

What systems do I need to run usage-based pricing?

At minimum: reliable usage data capture, and automated billing logic that can handle variable consumption. In practice, many companies underestimate this operational complexity. The 2026 PwC and m3ter survey found that 72% of software leaders with usage pricing lack confidence in their exposure to revenue leakage - a significant revenue integrity risk that modern monetisation infrastructure is designed to address.

Should we switch to usage-based pricing if we currently charge per seat?

Not necessarily. Per-seat pricing still makes sense where software delivers discrete value to individuals - collaboration tools, learning platforms, personal productivity software. The better question is whether a usage metric exists that more accurately reflects the value your customers derive. If it does, a hybrid model - retaining a subscription element while introducing usage-based components - is often the right first step, rather than a full switch.