Last month, we gathered 13 senior leaders from the world’s most sophisticated usage-based businesses, including Anthropic, Confluent, NVIDIA, HubSpot, and Notion, for our 2026 customer advisory board.

While the setting was serene, the discussions were urgent. It’s clearer than ever that we are—yes, yet again—at a major inflection point in monetization. As we transition from the Access Era of seat-based software to the Value Era of more outcome-focused services, we’re seeing a fundamental shift. Software isn't just a tool humans log into anymore. Now, it's a service that performs autonomous work. This requires moving past the idea of billing as a utility and toward a more sophisticated framework of monetization intelligence.

Which pricing era are we in now? Get oriented with the Monetization Operating Model whitepaper.

As companies prepare for an agent-led economy, here are the biggest strategic takeaways from our sessions.

Machine as customer: Preparing for agentic monetization

One of the more future-looking themes from these discussions was the recognition of the rise of agentic monetization. Leaders we spoke to are already preparing for a world where AI agents (not humans) are the primary consumers of API-driven services.

In this reality, pricing must transition from human-readable to machine-actionable. Traditional pricing was designed for human understanding and psychology. "Good/Better/Best" packages or high price anchors were established because of foundational research on how the human brain processes information and calculates decisions. But agents evaluate thousands of variables simultaneously to optimize for price, latency, and performance in real time, likely making those tiers moot. 

Think of a consumer agent booking a vacation: it can review a plethora of sites simultaneously to provide an informed perspective. In a business context, that agent will bring this same computational scale to evaluating service providers, likely making it less a matter of choosing between plans and more a matter of the agent making sub-second tradeoffs between cost, latency, and performance to function effectively. 

This fundamentally changes the commercial interaction. Rather than users choosing from plans, the world is evolving toward budget allocations, where a human sets priority, criteria, and budget, and the agent autonomously manages spending decisions. This requires a strategic shift from static monthly invoices to real-time primitives. By exposing billing data via APIs, companies allow agents to manage their own budgets, monitor credits, and make autonomous purchasing decisions based on the value of the task at hand.

The CFO’s new mandate: Margin as a weapon

Growth was often pursued at any cost in the early days of SaaS, and it often was a winning move. But with the high overhead of GPUs and LLM tokens, that era is over. When a single user action can fan out into dozens of downstream model calls and tool invocations, cost becomes highly variable. Similarly, without real-time visibility, profitability becomes a guessing game. Throughout the sessions, leaders shared that their finance teams are now desperate for real-time margin visibility.

What that translates to is that margin is now becoming a weapon. In the AI era, monetization has become a systems problem, where revenue observability—the ability to trace an event from usage to recognition in one click—is a strategic differentiator. Billing systems must provide the financial controls necessary to decide which workloads are profitable in real-time. This isn’t just about shut-off valves but is more about intelligent enforcement and having the data to decide whether to route a request to a cheaper model or allow a customer to briefly go into the negative based on their lifetime value.

As one attendee put it: 

“Billing is a mask for what the actual need is. What is the value?”

The end of ‘simple’ pricing

A common refrain we’ve all heard is that pricing must be simple to be effective. But these discussions had members arguing firmly that complexity is actually becoming a competitive advantage.

While humans crave simplicity, agents thrive on granularity. More levers let companies align revenue even more precisely with the diverse costs of AI, differentiating between async processing, real-time reasoning, or specialized model outputs. To capture revenue accurately, companies are moving toward multivariable models. For example, some are charging differently for asynchronous processing versus real-time reasoning because it allows them to align revenue directly with costs and the value delivered.

To manage this, many are turning to a universal credit model, which lets companies offer a simple, fungible credit balance to the end user, while abstracting the underlying pricing mechanics that are sophisticated enough to reflect the true, complex cost of service.

The need for an orchestration layer

As monetization strategies become more sophisticated, the underlying infrastructure also has to evolve. With this evolution comes the rise of monetization engineering as a discipline, where monetization is treated like production infrastructure: resilient, observable, and built for constant iteration. The leaders we heard from in these discussions highlighted a growing need for a logic layer in the revenue stack, meaning a system that can bridge the gap between real-time product usage and legacy financial systems like NetSuite, and CRM/CPQ tools or traditional payment processors.

The consensus was that the modern stack requires three specific capabilities to succeed:

• Operational nimbleness
  Pricing is now a core product feature. Infrastructure must give teams the ability to iterate on and ship new pricing models quickly—in days or weeks, not quarters—to keep pace with the AI market.
• Structural flexibility
  Modern billing has to evolve from ‘record-keeper’ to ‘runtime system,’ where it continuously computes pricing and entitlements as events happen, rather than in batch jobs. It must act as the connective tissue that can ingest massive volumes of data and apply granular logic that traditional billing tools aren't built to handle.
• Strategic evolution
  Ownership of monetization is becoming a centralized, cross-functional monetization operating model. Technology partners have to move beyond being mere vendors to becoming design partners that help define how the ‘agents as operators’ concept will eventually manage the revenue stack itself.

Leading the next wave of monetization

Within the far-ranging discussions in this year’s customer advisory board, the clear thread was that the new standard for billing is defined by three things: flexibility, real-time processing and intelligence, and readiness for the emerging agent-led era.

Software’s Access Era is being replaced by the Value Era, where trust is built through transparency and predictability. If users don't understand what they're paying for, or why one query costs more than the last, the relationship breaks. At Metronome, our focus is on being the control plane for the world’s most sophisticated usage-based businesses. As the industry moves toward agent-led growth, we’re excited to continue partnering with our customers as the way value is captured and defended in the AI economy becomes better defined.