Why AI companies have adopted usage-based pricing

Usage-based pricing is transforming how software and AI services are monetizing their product. Unlike seat-based pricing where companies used to charge a set fee per user account regardless of actual activity, this model charges customers according to actual usage of a product.

Adoption of usage-based pricing among SaaS companies has accelerated rapidly in recent years. According to a survey, the percentage of SaaS companies using any form of UBP rose from 30% in 2019 to about 85% by 2024.

But what has driven this surge in adoption? Let’s break it down:

1. Usage is Tied to Infrastructure, Not User Count:
   AI models, especially those using deep learning, are computationally intensive. The primary cost drivers are not the number of users, but the GPU cycles, API calls, and tokens processed. Unlike a traditional SaaS tool where user count and active sessions are the main cost factors, AI companies face steep infrastructure bills that can scale unpredictably. In fact, infrastructure costs for some AI businesses have risen from just 10% of their cost to as much as 35-40% as they scale, making efficient cost recovery through usage-based pricing essential.
   
   

   

   
   

2. AI Value Doesn’t Scale Per Seat:
   The value derived from an AI product is often not tied to the number of users but rather the volume of work being processed. For example, one customer might run a few hundred model inferences per month, while another could be processing millions of requests. In both cases, the number of “seats” might be identical, but the compute load and associated costs can be vastly different. This makes seat-based pricing a poor fit for AI workloads.
   
   

   

   
   

3. Fair and Transparent:
   Usage-based pricing directly ties cost to value. Customers pay in proportion to the benefit they receive. This alignment builds trust and reduces friction. In contrast to flat subscription fees, where customers might pay for features or capacity they never use, UBP ensures that if they consume less, they automatically pay less. This perceived fairness encourages customers to start small and scale as needed.
   
   

4. Lower Barrier to Entry for New Customers:
   Usage-based pricing lowers the risk for new customers by eliminating large upfront commitments. This makes it easier for early-stage projects or smaller teams to start using a product without overcommitting. For companies trying to penetrate new markets or onboard small but promising customers, this can be a critical advantage.
   
   

5. Best-in-Class Retention and Upsell Opportunity:
   Successful customers naturally use more over time, and because pricing scales with usage, their spending can grow without a heavy upsell effort. Studies show that companies with UBP often report higher net dollar retention (NDR) compared to those with pure subscription models. This makes it a powerful tool for increasing customer lifetime value without the need for constant re-negotiation.

While the tech industry at large has embraced the concept of usage-based pricing as the direction forward, especially for cloud services, APIs, and anything usage-intensive. But, there is a hesitancy is customers around predictability. Not all customers are comfortable with variable bills. Especially in enterprise IT procurement, there is often a preference for known costs that can be budgeted annually. If usage (and thus cost) can swing widely, some customers get uneasy or demand caps. There have been cases of sticker shock, e.g. a team accidentally overuses a service and gets a much larger bill than expected, leading to complaints.

Beyond just cost predictability, purely usage-based pricing introduces several operational and financial complexities for AI companies. Budgeting becomes a headache for CFOs, as forecasting variable costs is inherently difficult. Revenue recognition, especially for public companies, can be tricky, requiring real-time data pipelines and precise accounting controls to avoid regulatory issues. Also, unpredictable bills can lead to customer churn if the perceived value doesn’t match the cost, increasing the risk of lost accounts.

This is why AI companies are experimenting with different pricing models strategies and are also introducing measures like cost alerts, usage caps, and prepaid plans (also known as credits) to build confidence. For example, OpenAI introduced prepaid credits that customers can buy upfront, offering a way to manage expenses while still benefiting from a pay-as-you-go model. Similarly, cloud providers often offer spend limits to reduce anxiety around unpredictable costs.

Few models which AI companies have recently tried to adopt include:

• Freemium model: Buffer gives free users access to AI Assistants to drive deeper engagement - and ultimately, more upgrades.

• Free Trial: Intercom offers 10 free AI tickets/user/month to let users taste the value, without cannibalizing paid plans.

• Personal Add-On: Notion sells AI access separately for $8/mo, turning free users into revenue without forcing a full plan upgrade.

• Workspace Add-On: Slack charges for AI at the workspace level, doubling revenue for many accounts without offering free trials on smaller plans.

• Fixed Rate Usage: Algolia sets a flat rate for AI Recommendations, giving customers a low-risk way to experiment with new functionality.

• Credit Model: Airtable uses credits for AI usage - included in plans, but with easy options to buy more in bulk.

• Premium Tier: Loom created a new 'Business + AI' tier - smart move when an add-on wouldn't fit their existing free/paid structure.

• Tiered Usage Model: Miro offers AI features across all plans, but limits usage via credits - perfect for driving upgrades from power users.

• Controlled Rollout: Asana launched AI Studio only for select Enterprise/Annual customers - allowing them to test and learn before scaling.

Each of these approaches targets a different stage of the customer journey, allowing companies to match pricing with perceived value.

But implementing these pricing models is hard:

Companies do not often stick to a single approach to monetize their product. These strategies evolve over time. What starts as a simple sprint to add usage-based billing quickly turns into an endless list of development tasks. You’ll find yourself needing to support real-time metering, dynamic pricing, hybrid models, and complex entitlements as customer demands grow.

Moreover, margins are thin in AI, and every second of latency in your billing pipeline can mean lost revenue or worse—overcharging customers and breaking their trust. Complex pricing models like credits, tiered usage, or hybrid plans add even more strain on your architecture. We’ve tried to elaborate more about the engineering complexities in implementing usage-based pricing model in this blog.

At times, even the conventional billing tools like Stripe Billing or Chargebee falls short. While these tools were powerful for conventional SaaS, it wasn’t originally designed for the demands of real-time, usage-driven AI businesses. For companies experimenting with pricing, enforcing feature access, or managing prepaid credits in real time, these solutions quickly becomes a bottleneck. You end up writing endless scripts, building custom integrations, and maintaining complex middleware just to make them work for AI pricing.

That’s why Flexprice exists. We make sure AI companies don’t just build cool tech but actually make money from it. Open-source, usage-based, and developer-friendly because pricing shouldn’t slow you down.

So yeah, you could keep hacking Stripe Billing or Chargebee to fit your AI pricing model or you could just use Flexprice. But hey, you do you. 😏

Ready to take control of your pricing?
Try Flexprice today. It’s open source, free to use, and built for builders like you.