Skip to main content
Be The Investor
BTI Market Newstechnology
Back to News

Salesforce's AI Usage Is Growing Faster Than Its Billing Model

Salesforce's AI Usage Is Growing Faster Than Its Billing Model

Salesforce's AI adoption is visible in usage and contract growth, but the investment case depends on converting flat enterprise licenses into durable consumption revenue at renewal.

Salesforce has evidence that customers are using its AI products, but the harder question is how quickly that activity becomes recurring revenue. The distinction matters after a sharp share-price recovery and as investors debate whether AI will strengthen or weaken the company's pricing power.

Salesforce reported fiscal second-quarter 2027 revenue of $11.35 billion and constant-currency current remaining performance obligation growth of 14%. Agentforce and Data 360 annual recurring revenue reached nearly $3.9 billion, up more than 210% from a year earlier.

Usage is not the same as monetization

At Salesforce's September investor session, management highlighted heavy Agentforce use among its top customers. The source article reported that annual recurring revenue for the top 100 customers by Agentic Work Units had roughly doubled since Agentforce launched.

That cohort is selected for high usage, so it should not be treated as the average customer. It does show that customers can expand spending when agents become embedded in sales, service and internal workflows.

The timing problem comes from contract design. Large customers often sign Agentic Enterprise License Agreements, or AELAs, that bundle broad access for a fixed price. These agreements can accelerate adoption but delay the point when additional usage produces additional revenue.

Renewals will reveal pricing power

Salesforce's $1.6 billion, three-year agreement with the U.S. Department of Veterans Affairs demonstrates the scale that an AELA can reach. Its expanded Siemens partnership also shows that Agentforce is moving into operational workflows, including lead qualification for thousands of sellers.

The investment case now turns on renewal economics. If flat licenses convert into contracts priced around usage or measurable outcomes, AI adoption can lift revenue without requiring a new customer relationship. If customers resist repricing, usage may grow faster than revenue for longer than bulls expect.

Current remaining performance obligation is the cleanest near-term test because it measures contracted revenue expected within 12 months. Sustained growth near the latest 14% rate would suggest adoption is becoming commitments. A slowdown would indicate that product enthusiasm has not yet translated into stronger billing.

Find us onGoogle News

Latest comments

BT
Community Guidelines