Summary

  • Okta says accelerating the shift of professional services to partners will reduce fiscal-2027 total-revenue growth by approximately one percentage point.
  • In Q2, Professional services and other produced US$12 million of revenue against US$19 million of cost. Okta's presentation reports a negative 53.4% non-GAAP professional-services gross margin.
  • Moving the work can improve Okta's reporting perimeter without proving that customer deployment cost, lead time or execution risk has fallen. That requires a direct-versus-partner delivery receipt.

Okta put an unusual concession inside an otherwise conventional fiscal-Q2 outlook. Its full-year revenue guidance already includes an approximately one-percentage-point drag on growth caused by a management decision: professional-services work is moving to partners more quickly.

That is not automatically a demand warning. Subscription revenue grew 12% to US$793 million, while total revenue reached US$805 million. RPO, which Okta calls subscription backlog, rose 17% to US$4.858 billion; the portion expected over the next 12 months rose 14% to US$2.585 billion. Professional services and subscription demand occupy different ledgers.

The reason for accepting lower reported revenue appears one line further down the statements. In the Form 10-Q, Professional services and other generated US$12 million in the quarter and cost US$19 million to deliver. The subtraction is a US$7 million GAAP gross loss, or a calculated negative 58.3% margin. Okta does not publish that GAAP percentage.

Its investor presentation supplies the adjusted version. After adding back US$1 million of stock-based compensation in the cost line, non-GAAP professional-services gross profit was still negative US$6 million. The reported non-GAAP margin was negative 53.4%.

The minus sign therefore belongs to a delivery function, not to the subscription platform as a whole. Okta reported US$648 million of GAAP subscription gross profit and US$641 million of consolidated gross profit. A small service line was consuming more direct cost than the revenue it recorded.

A growth headwind is not a demand receipt

The one-point disclosure is easy to overread. It refers to the effect on fiscal-year total-revenue growth. It is not one percent of guided revenue, an exact dollar reduction, or a disclosed estimate of lost subscriptions. Okta did not publish the counterfactual revenue it would have recognized had the work remained in-house.

The line is already shrinking. Professional-services-and-other revenue fell from US$17 million to US$12 million year on year, while its GAAP gross loss widened from US$4 million to US$7 million. For the first six months of fiscal 2027, revenue was US$27 million and cost was US$39 million, a US$12 million gross loss. A year earlier, US$32 million of revenue and US$40 million of cost produced an US$8 million loss.

That persistence matters. The decision is not justified by one anomalous project in the public accounts. But the filings do not say which consulting, migration, configuration, integration, training or other work is leaving, which regions or customer cohorts are affected, or how quickly the transfer will occur.

Nor do they isolate the margin benefit. Okta guides to a 26% non-GAAP operating margin and a 28%-29% free-cash-flow margin for the year. Those are consolidated outcomes. Without a cost bridge, the one point of revenue-growth headwind cannot be exchanged mechanically for a known operating-margin gain.

The work moves; it does not disappear

Software implementation has a physical economics even when the product is cloud-delivered. Someone scopes the environment, maps identities and permissions, connects applications, migrates policy, tests failure states, trains administrators and repairs exceptions. The associated labour can sit inside Okta, inside a systems integrator, inside the customer, or across all three.

Moving that labour to partners can be rational. A partner network may provide more certified staff, local presence and knowledge of adjacent systems. Okta can concentrate capital and management attention on subscription software while an integrator earns the implementation revenue. If capacity expands and deployments finish sooner, a smaller services line can support a larger subscription base.

But no cost is extinguished merely because it leaves Okta's income statement. A partner must pay staff and recover a margin. The customer may pay that invoice directly, making the expenditure invisible to Okta's revenue. Okta may still bear presales, enablement, support, escalation and remediation cost. The relevant economic perimeter is therefore the complete deployment, not one vendor's gross margin.

The transfer also changes control. Okta controls product architecture, partner qualification, documentation, sales incentives and escalation interfaces. Partners control staffing, utilisation, project planning and much of day-to-day execution. Customers control internal readiness and ultimately bear coordination delay. The public record does not allocate every contractual duty, so none should be invented.

Subscription value arrives after implementation

The commercial chain is longer than a licence signature. A subscription is sold; a deployment scope is accepted; qualified labour becomes available; the product reaches a useful production state; defects and exceptions are resolved; and the customer renews or expands.

Partner delivery succeeds economically only if it improves that chain, or at least preserves it at lower system cost. A fast sale followed by a long implementation queue can delay adoption. Poor configuration can create support demand or security exposure. Expensive partner work can make the software's total cost less attractive even when Okta's own margin improves.

The opposite is equally possible. A specialist partner may deploy faster than a small internal team, integrate complementary products and place variable project cost with the party best equipped to manage it. The current disclosure establishes neither outcome.

RPO and cRPO help measure contracted subscription value, but they do not reveal who delivered the project, what the customer paid for it, when the installation became useful or whether rework occurred. Professional-services revenue becoming smaller is therefore not itself proof of an asset-light success.

The missing partner-delivery bridge

A useful receipt would compare Okta-led and partner-led work by service type, region and customer cohort. It would show project count and value, certified capacity, attach rate, waiting time, time to production, completion, rework, support escalation, customer total cost and subsequent renewal or expansion. The financial bridge would reconcile revenue relinquished by Okta with internal cost avoided and partner or customer cost created.

Without that bridge, the disclosed decision remains legible but incomplete. Okta is reducing exposure to a service line with a deeply negative margin. That can make the company more scalable. It can also push an essential part of time-to-value into a network whose economics and quality are less visible. The difference will appear in deployment receipts, not in the disappearance of revenue alone.

Sources