An AI agent that resolves a customer problem in three steps can generate less revenue than one requiring ten—if its provider charges per action.

That paradox exposes a growing weakness in subscription economics: the amount customers pay is increasingly disconnected from the value they receive.

For investors, the implications extend beyond pricing strategy. Annual recurring revenue (ARR), customer growth, and retention have long served as indicators of software business quality. But AI can simultaneously increase customer productivity, reduce the number of paid licenses, and introduce substantial variable computing costs.

A company can therefore deliver a better product without generating proportionately higher revenue—or profits.

The emerging investment question is not simply how much recurring revenue a business generates, but how effectively it converts customer value into sustainable gross profit.

From Selling Access to Selling Results

Traditional enterprise software scales through user licenses. More employees generally mean more subscriptions.

AI disrupts that relationship. When autonomous agents perform work previously handled by employees, customers can increase productivity while reducing their need for software seats.

Salesforce's Agentforce illustrates the transition. Its pricing includes $2 per conversation and a Flex Credits model that can charge approximately $0.10 per action.

Intercom takes a different approach. Its Fin AI Agent charges $0.99 per customer-service resolution rather than per interaction.

The distinction matters.

Salesforce's action-based model monetizes activity. Intercom's resolution-based model attempts to monetize results.

Under activity-based pricing, a less efficient agent completing more operations can produce more billable revenue. Outcome pricing potentially reverses that incentive.

But it introduces another problem: who determines whether an outcome has actually been achieved?

Intercom can classify an interaction as resolved when a customer accepts an answer or leaves without requesting further assistance. Neither necessarily proves that the underlying issue was fully addressed.

This is a classic contracting problem: when performance is difficult to verify, companies can optimize for the measurement rather than the customer's actual objective.

For investors, outcome-based pricing deserves a premium only when the vendor can demonstrate reliable measurement, low dispute rates, and attractive margins after accounting for unsuccessful attempts.

Netflix Demonstrates a Different Model

Consumer subscriptions face a separate challenge: customers increasingly treat recurring payments as temporary commitments.

Deloitte's 2026 Digital Media Trends report found that 41% of U.S. consumers had canceled at least one streaming subscription over six months, while 22% had canceled and later returned to the same service.

Netflix has responded in part by expanding advertising.

Its advertising revenue exceeded $1.5 billion in 2025, more than doubling year over year. Yet advertising still accounted for only approximately 3% of its $45.2 billion annual revenue.

The strategic significance lies less in advertising's current contribution than in Netflix's ability to monetize one audience through two paying parties: subscribers purchasing access and advertisers purchasing attention.

This broadens revenue opportunities without requiring equivalent increases in subscription prices.

For investors, however, it also changes the risk profile. Subscription revenue depends primarily on subscribers and pricing; advertising revenue introduces exposure to marketing budgets, advertiser demand, and the effectiveness of audience monetization.

Revenue diversification can improve monetization without necessarily improving revenue predictability.

Why Hybrid Pricing Is Gaining Ground

A Stripe survey found that 56% of AI company leaders were already using hybrid pricing models, combining subscriptions with usage-based charges.

The economic rationale is straightforward.

Fixed subscriptions provide predictable revenue but can expose vendors to disproportionate servicing costs from heavy users. Usage-based pricing better reflects consumption but creates spending volatility for customers. Outcome pricing aligns payment with results but transfers performance and measurement risk toward the provider.

Hybrid contracts can separate these functions: fixed fees cover access and availability, variable fees recover incremental consumption costs, and performance payments capture a share of measurable value.

The important innovation is not billing flexibility itself. It is the ability to distribute economic risk more efficiently between customers and providers.

That distinction has direct valuation consequences.

Two companies with identical ARR could have fundamentally different economics if one earns high-margin subscription revenue while the other depends on costly, volatile AI workloads.

Investors consequently need to examine gross profit per task, cost-to-serve variability, and the relationship between customer productivity and vendor revenue—not just headline growth.

The Next Source of Pricing Power

Monthly subscriptions are unlikely to disappear. They remain effective for maintaining access and securing predictable cash flow.

But as software increasingly performs work rather than merely enabling it, the traditional unit of monetization becomes less reliable.

The competitive advantage will belong to companies that can identify measurable customer value, capture an economically defensible share of it, and retain attractive margins as delivery becomes more efficient.

The next subscription economy will not be defined by how often customers pay, but by how much of the value created a business can sustainably capture.

References