The Audit

2026-08-04

When more services online is not the same as benefits realised

Service Victoria had more services online and high satisfaction among users who completed transactions. But the Victorian Auditor-General’s Office found in its 2026 follow-up that later activity measures could not show whether the platform was delivering the transaction savings set out in its original business case.

Spring Street government buildings in Melbourne at night.
Image adapted from “Spring St Government buildings at Night Dec 2012” by Nick-D, via Wikimedia Commons, licensed under CC BY-SA 3.0. Cropped and resized.

Service Victoria is the Victorian Government's digital service platform. Its 2015 business case was built around moving government transactions onto a common platform and reducing service-delivery costs. The Victorian Auditor-General's Office (VAGO) recorded that the case predicted an annual financial benefit of $61 million, growing to between $120 million and $150 million a year by 2020 if all Victorian Government transactions moved to Service Victoria.

VAGO first reported on Service Victoria in 2021. It found that the platform had improved customer experience by moving some services online, but had not reduced transaction costs as intended. In 2026, VAGO returned through a limited-assurance follow-up that asked whether agencies had implemented their action plans, and whether customer experience had improved and transaction costs had reduced since the earlier audit.

The platform had expanded with VAGO recording 17 end-to-end transactions in June 2020 increasing to 51 in April 2026. But the 2026 follow-up also found that the Department of Government Services still could not show whether Service Victoria was delivering the benefits intended in the 2015 business case, or how many transactions it delivered.

The reported measure changed from transaction cost to activity cost. The business case and the 2021 audit were concerned with transactions and transaction costs while later reporting used activities and activity costs. VAGO said the activity-cost measure could not be compared with the 2021 audit because it was not a cost-per-transaction measure. It also said Parliament and the public could not tell from the later measure whether transaction costs had reduced.

VAGO had recommended measures for transaction cost, transaction volume and the number of services delivered because only customer satisfaction was publicly reported. The 2026 follow-up said the Department of Premier and Cabinet did not implement those measures and the Department of Government Services did not adopt them. VAGO also said activity counts could give an inflated view of performance because they included actions that did not complete a transaction, such as viewing a digital card, proving or reusing identity, or opening a push notification.

Customer satisfaction measured a different part of the service. VAGO said satisfaction feedback from customers who had completed transactions remained consistently high. Customers who did not complete a transaction were not offered a satisfaction rating.

The later benefits reports did not restore a transaction-cost measure. VAGO recorded a 2020-21 internal benefits report showing $35.7 million in benefits which was $18.6 million below target. It recorded a 2021-22 interim report showing $35.5 million in benefits under a revised methodology, with no comparison to a target. VAGO said the Department of Government Services had not produced interim or final benefits measurement reports since 2022. It also said a proposed framework completed in 2022 was never endorsed or put to formal use.

The 2015 forecast also depended on more government transactions moving onto Service Victoria. VAGO said the Department of Government Services did not have a clear strategy for Service Victoria's future or for encouraging agencies to transfer services to the platform. It also said the departments did not know how many transactions were delivered across government, digitally or otherwise. VicRoads registrations and licensing services and land title transactions were among the high-volume transactions VAGO said had not transitioned by 2026.

Other digital-service guidance uses similar measurement language. GOV.UK guidance defines cost per transaction as the cost to government each time someone completes the task a service provides, and tells teams transforming an existing service to measure it early as a baseline against future data. Digital NSW uses cost per transaction, user satisfaction, completion rate and digital take-up as service measures. The Australian Government Digital Service Standard tells agencies to establish a baseline and identify performance indicators. Queensland's Digital Service Standard also refers to user satisfaction, digital take-up, completion rate and cost per transaction where possible.

Service Victoria had more end-to-end transactions available, and satisfaction among completing users remained high. But VAGO said the later activity-cost measure could not be compared with the 2021 audit, and did not show whether transaction costs had reduced. That is the part of the original business case the 2026 follow-up still could not demonstrate.

Sources