Answer capsule
Workday reported on August 27 that more than 5,500 customers now use at least one of its organic agents. That is a provider-reported customer count, not evidence that a particular finance workflow is used deeply, produces a financial return, or preserves the controls a CFO needs. A buyer still needs a workflow-level baseline, complete cost record, consequence test, and reconciled outcome before treating agent adoption as value.
What the source establishes
- Workday's August 27, 2026 fiscal 2027 second-quarter release says more than 5,500 customers now use at least one of its organic agents.
- The same release says the customer count was up more than 35% from the prior quarter and that AI drove more than 25% of new annual contract value; both are provider-reported company-level metrics.
- Workday separately introduced Adaptive Decision Intelligence for finance and operations and said its Financial Audit Agent became generally available.
- The release does not provide customer-level use intensity, workflow baselines, realized benefits, control outcomes, total costs, or a method that attributes financial results to any one agent.
Keep the announced populations separate
The CFO should preserve exactly what Workday reported and what it did not. The 5,500-plus population is customers using at least one organic agent, not 5,500 finance departments, not a count of active finance users, and not a cohort shown to have realized value. The more-than-35% quarter-over-quarter change describes that provider-reported customer count. The statement that AI drove more than 25% of new annual contract value is a commercial metric at Workday, not a customer return measure. Adaptive Decision Intelligence and the generally available Financial Audit Agent are separate product statements. A decision record should keep customer count, commercial contribution, feature availability, workflow use, and buyer outcome in different fields so a persuasive earnings narrative cannot silently become local evidence.
Name the finance job and its baseline
A value case begins with one owned finance job: for example, preparing an audit evidence package, explaining a forecast variance, or modeling a specific scenario. Record the current inputs, responsible roles, review and approval steps, elapsed and hands-on time, error and rework rates, exception volume, control evidence, downstream posting or disclosure effect, and the consequence of a wrong result. Then state what the agent may draft, retrieve, calculate, recommend, or initiate and what remains a human determination. A broad customer adoption count cannot answer whether the buyer has the required entitlement, connected records, configured policy, qualified reviewer, or representative volume. It also cannot establish whether saved preparation time is redeployed, whether control work shifts elsewhere, or whether a faster draft creates downstream reconciliation cost.
Build a complete cost-and-consequence case
The finance case should include subscription and feature entitlements, implementation, data preparation, integration, model or consumption charges, testing, controls, monitoring, training, support, exception handling, remediation, and the cost of parallel operation during changeover. Benefits should be tied to completed and accepted work: avoided external effort, reduced controllable rework, shorter decision latency that changes an action, improved cash or working-capital results, or better audit readiness without a control loss. Test the risky tails rather than only the easiest cases, including incomplete evidence, changed master data, stale policies, conflicting sources, unsupported assumptions, unusual entities, currency and tax differences, and a reviewer correction. A benefit remains provisional until the corresponding financial record, approval, posting, disclosure, or audit package is reconciled.
Scale only from attributable operating evidence
Run a bounded comparison against the prior process and preserve cohort definition, assignment method, time window, volume, exclusions, failure and abandonment, reviewer effort, corrections, incidents, and total cost. Segment by workflow, entity, materiality, exception complexity, and user role so frequent low-consequence activity does not hide weak performance on consequential work. The decision owner should be able to connect a configured agent version and governed workflow to each accepted outcome, then show what changed after deployment and what plausible alternatives remain. If the evidence supports use, approve a limited scope with thresholds, stop conditions, review cadence, and a named owner. If it supports only adoption, describe adoption. Workday's aggregate customer and commercial statements are useful market signals, but they do not substitute for the buyer's own finance evidence.
Turn this source into a reviewable decision
For AI for CFOs, use this briefing as a dated decision record rather than a substitute for the source. Preserve Workday Announces Fiscal 2027 Second Quarter Financial Results, the exact URL, the August 27, 2026 review date, the supported facts above, the editorial interpretation, the limitations, and any buyer-specific evidence. Link that record to the decisions most directly affected: Planning and scenario analysis; Management reporting and external disclosure support; Internal control and audit evidence; Finance policy and self-service. State whether the source changes the scope, evidence requirement, control, sequence, or only the language used to describe the decision.
Before action, name the accountable owner, affected population and workflow, exact offering or configuration, source data and rights, human decision point, exception and appeal path, complete cost, expected benefit, failure and stop conditions, retained evidence, and next review date. Keep official facts, provider statements, buyer observations, representative tests, measured outcomes, editorial inferences, and unknowns visibly separate. Reopen the record when the source, offer, model, integration, data, policy, population, responsible person, or measured result changes.
Limitations and unknowns
Workday is the provider source. Its August 27, 2026 fiscal-results release reports that more than 5,500 customers use at least one organic agent, that the count increased more than 35% from the prior quarter, and that AI drove more than 25% of new annual contract value; it also announces finance-related capabilities. It does not independently establish the customer-count method, use frequency or depth, finance-user population, feature-level attribution, buyer entitlement and configuration, workflow fit, input quality, accuracy, exception behavior, human effort, control effectiveness, total cost, time redeployment, cash or accounting effect, or realized customer value. Current contracts and documentation, configured-workflow records, complete costs, representative tests, reconciled finance outcomes, and qualified finance, accounting, audit, tax, privacy, security, procurement, accessibility, and legal review control.
Decision test
Ask whether the source changes the decision itself, the evidence required, the implementation sequence, or only the language used to describe an existing capability. Record which claims are directly supported, which are provider statements, which require an independent test, and which remain unknown. A source-linked review should make uncertainty easier to see, not bury it inside a blended score.
Questions to take into review
- Which planning model and dimensions ground the answer?
- Can every assumption be traced to an owner and date?
- Which source supports each number and assertion?
- How is materiality assessed outside the model?
- Is the AI itself in scope for change and access controls?
- Can evidence provenance survive export and retention?
- Which documents are authoritative and effective today?
- What topics always require a person?
The publication supports research and executive decision preparation. It does not provide legal, financial, accounting, employment, clinical, cybersecurity, investment, procurement, or implementation advice.