HM Treasury will not confirm whether it is joining the Matrix shared-services programme until December 2026 — a decision originally expected this summer — because the feasibility documents needed to evaluate the migration from its current Oracle Fusion environment have not yet been delivered (The Register, 11 June 2026). The knock-on effect: the programme’s go-live target, previously set for 2028, has slipped to 2029.
The Decision That Is Not Happening Yet
Matrix is a government-wide ERP consolidation contract worth £144.3 million, awarded to Workday (SaaS finance and HR) and Cognizant (systems integration). It sits within the broader £1.7 billion Shared Services for Government Strategy, which Treasury has co-funded to the tune of £1.15 billion since 2021 — without having agreed to migrate its own systems.
Nine departments are in scope: Cabinet Office, DSIT, DESNZ, DCMS, DBT, AGO, DfE, DHSC, and HM Treasury. If Treasury and DfE both remain outside the programme, the projected 15-year benefits fall from £185 million to £109 million in net present value terms, according to the programme’s own sensitivity analysis.
The summer 2026 decision window was the first concrete opportunity for Treasury to formally join. It has now passed without a decision. The next checkpoint is December 2026, contingent on feasibility documentation arriving before the end of summer.
Why the Feasibility Documents Are Not Ready
Treasury’s current platform — Oracle Fusion, heavily customised to manage the financial workflows of a sovereign treasury — presents a specific migration challenge. Workday, as deployed under Matrix, is a standardised SaaS configuration. The gap between “what Treasury does today” and “what the standard Workday instance supports” has not yet been formally quantified.
That quantification is precisely what the pending feasibility documents are supposed to establish: what functionality would be lost, what custom processes would need to be re-engineered or accepted as gaps, and at what cost. Without it, Treasury cannot make an informed decision — and is not making one.
This is not unusual in large transformation programmes, but it creates a sequencing problem: the programme advances at pace while one of its anchor tenants is still evaluating whether to board at all.
Three Dynamics CIOs Should Recognise
Customisation as migration debt. Oracle Fusion in a heavily bespoke configuration is not a standard Oracle Fusion migration problem — it is a custom-system migration problem that happens to run on Oracle Fusion. Every bespoke process, every report built outside the standard configuration, every workflow tweaked over the years is a decision point in the migration. The longer the organisation has had the system, the larger the inventory of those decisions. Treasury’s situation is an extreme example of something most large-enterprise CIOs manage at smaller scale.
Missing feasibility documentation as a decision-stopper. A group-level or multi-entity ERP programme that produces a consolidated business case without individual-entity feasibility assessments creates blind spots. Treasury is blocked not because the overall programme business case is weak, but because the specific analysis for its own environment does not exist yet. For any CIO governing a multi-entity transformation, the lesson is direct: each entity needs its own migration assessment before it can make a credible commitment.
Cross-entity dependency risk. The programme’s financial model rests on a critical mass of participating departments. Treasury’s delay creates uncertainty for the rest of the cluster. The departments that have committed are now running on a programme timeline (2029) that is already a year later than the original plan, partly because one key anchor tenant has not yet decided to join. This is the shared-services version of the group ERP deployment where a major subsidiary’s late commitment destabilises the rollout sequence for everyone else.
What Happens Next
Two dates govern the next phase:
End of summer 2026: feasibility documentation due. If this slips further, the December 2026 decision is at risk. Every additional delay shrinks Treasury’s window to join the programme on its current configuration rather than inheriting a Workday setup that has already been locked in by the other eight departments.
December 2026: formal join/no-join decision. If Treasury confirms participation, a Treasury-specific migration roadmap from Oracle Fusion to Workday still needs to be designed — on top of an already-live programme. If Treasury declines, the programme loses scope, net benefits are recalculated downward, and DfE’s own decision gains more weight.
The Department for Education is in a structurally similar position. Between the two of them, they hold most of the programme’s remaining swing in financial returns.
For context on the UK public-sector ERP landscape, see our analysis of the UK ERP market: Sage, Access Group, and Making Tax Digital and our piece on Making Tax Digital Income Tax coming into force in April 2026.
