Workday grew subscription revenue 13.9% on the pricing model agents were meant to replace

The administrator who switched on a Workday agent last quarter did not get a new line on the invoice. The capability showed up inside a subscription still renewed by seat, and Workday's own language gives that away. Co-founder and CEO Aneel Bhusri says AI drove more than 25% of new annual contract value ^[4]. That figure comes from an internal attribution rule Workday applies to classify deals as AI-influenced. Someone inside Workday decides which deals count as AI-influenced, and no customer signed a contract that reads "agents, $X per month."

The adoption figure has a load-bearing qualifier in it: customers using *at least one* organic agent ^[5]. Breadth is the cheapest number for a suite vendor to move, because switching on one agent in one process, for a customer who already owns the suite, requires no procurement cycle at all. It does not tell a buyer how much work routes through an agent in a week, or whether the Financial Audit Agent actually shortened anyone's evidence packages after it went generally available ^[15].

The operating leverage underneath is real. Non-GAAP operating income rose 21.2% while total revenue rose 12.8% ^[18], and subscription revenue is 93.3% of the total ^[21], so the company's growth rate is essentially its seat-based subscription growth rate. One caveat on the earnings line: GAAP diluted EPS of $2.57 included a $1.52 per-share tax benefit from an intra-entity transfer of intellectual property ^[9]. Strip that and the GAAP figure is $1.05 against $0.84 a year earlier ^[23]: a solid quarter, well short of the tripling the headline implies.

The backlog split is where a renewal team should look. Back the near-term book out of the total on both sides of the year, and the committed revenue sitting beyond twelve months grew about 5.2%, roughly $18.4bn against $17.5bn ^[22], against 14.2% growth in the twelve-month book ^[10]. Customers are still buying. They are committing less far out than the headline suggests, which is what you would expect from a buyer who has been told for two years that agents will change the seat count and cannot yet say by how much.

For your own renewal, two questions matter: whether the agent capability is priced separately or bundled, and whether it reduces the headcount you would otherwise buy seats for. Bundled and headcount-neutral makes it a retention feature, and you should value it at zero in the negotiation. Bundled and headcount-reducing means your seat count falls and the vendor's contract value falls with it. Priced separately and headcount-neutral is a straight add to your bill, and worth resisting. Priced separately and headcount-reducing is the only cell where a per-agent line item serves the vendor's interest as well as yours, and it depends on the vendor being confident enough in the savings to trade seats for meters. Workday's pricing tells you where it thinks it is: the capability stays inside the subscription. The guidance stays at 13% subscription growth. Metering it later remains an option Workday hasn't exercised.