Government and shared services: a whole of government blueprint
PUBLIC SECTOR · SHARED SERVICES · OPERATING MODEL
The situation
A state government was carrying budget deficits and needed its back office to cost less and work better. Corporate and shared services across the sector were delivered in a highly fragmented and inconsistent way: thirteen departmental clusters ranging in size from around 1,000 employees to around 90,000, together accounting for roughly 270,000 people, served by a patchwork of in house providers built for a single agency, multi tenanted providers serving several, and outsourced arrangements. Maturity varied enormously. So did the degree to which corporate functions had actually been integrated at cluster level following recent amalgamations.
Eleven corporate functions were in scope, from finance and human resources through industrial relations, workplace health and safety, technology, procurement, governance and risk, records, property and fleet, and asset management. There was no common framework for deciding what should be consolidated, into how many providers, under what governance, funded how, or measured against what. Each cluster was therefore making its own decisions on its own evidence, at its own pace, and the sector was heading towards a set of incompatible answers. The central department commissioned a blueprint. Deliberately a framework rather than a detailed design, because detailed design imposed from the centre would not have survived contact with thirteen directors general.
Before
Thirteen clusters, fragmented provision
Widely varying provider maturity
No common activity definitions
Cost and headcount measured, little else
Expectation of a single transition
The work
Retained to write the blueprint and take it to the Directors General Executive Committee. Six strands:
Mapped the current state across all thirteen clusters and all eleven functions, establishing who was served by whom, at what maturity, and where the recent amalgamations had and had not been followed through into corporate integration.
Built the case for change on external evidence rather than assertion. A benchmark study of 193 public and private organisations across the United States, Europe and Asia Pacific showed that around six in ten shared service operations had achieved cost reductions above twenty per cent, with the large majority also reporting double digit improvements in quality, productivity and customer service. That was supplemented with documented outcomes from other state, federal and international governments running comparable reforms at comparable scale.
Designed the future state operating model. Five in house providers aligned to the largest clusters plus one multi tenanted provider for the remainder, with whole of government layers sitting above for wholesale technology infrastructure, property, fleet and contracting. The provider count was not arbitrary. It was drawn from the international precedent that had worked at similar scale, and sized against the actual employee distribution rather than the organisation chart.
Defined the functional scope in four layers, which is the discipline that stops these programmes overreaching. Transactional work first, being simple, repeatable and high volume, where economies of scale are real. Advisory work second, once performance had been demonstrated, where the economies are of skill rather than scale. Policy and control, and strategy, stay with the departments. Setting that boundary explicitly protects both the provider and the customer from the failure mode where a shared service centre is handed judgement work it was never built to do.
Set out the five conditions the reform would have to meet and designed each of them, not just named them: a governance structure running from a Cabinet committee through a directors general committee and provider boards down to functional operational committees; the service management layer between provider and customer that is routinely underinvested and is usually why these arrangements sour; a funding and pricing model that dealt honestly with operating cost, investment capital and benefit attribution; a benchmarking framework built on common activity definitions so that comparison was possible at all; and a staged implementation across three horizons rather than a single transition.
Developed a collaborative model to ensure there was a framework for clusters to participate and build on the design. Interviews with every cluster, every shared service provider and the central agencies, primary workshops, an interim report, formal written feedback from each cluster, and a senior officers feedback workshop before the final blueprint.
After
Six providers under one framework
Four layer functional scope
Common benchmarking taxonomy
Process level performance visibility
Three horizon staged transition
The outcome
A whole of government blueprint taken to the Directors General Executive Committee for endorsement, with a defined path to a change management and communications strategy, a reform programme office, and a detailed planning phase.
A common framework that let thirteen clusters make consolidation decisions that would still fit together, while allowing each to move at a pace matched to its own readiness rather than to a centrally imposed timetable.
A benchmarking foundation built on common activity definitions, giving whole of government visibility of total cost, cluster level visibility of efficiency and effectiveness against external targets, and agency level visibility of process cost and productivity. Without that layer, benefits realisation on a reform this size is an assertion.
A set of expectations stated plainly to the sector: that the full benefits of the end state would not be available during transition, that the transition would not be easy, and that it was pragmatic and achievable.