- 1. Building the plan
- 2. Ownership and the RACI
- 3. The cutover forum
- The cutover playbook
Every large program produces a RACI. Most are decoration. They get built in a workshop where nobody wants to name a single accountable party, so the compromise is to mark several people responsible and move on. The grid gets published, filed, and never reopened until something fails and somebody asks who owned it.
Cutover is where that bill comes due, because cutover runs outside business hours against a clock. At two in the morning, with a load failing and a gate approaching, someone has to be woken up. A cell that reads "Client" does not have a phone number.
Equal partners means no partners. Every row where both organizations are marked accountable is a row where the work has no owner and the contract has no evidence.
1Four numbers tell you whether a RACI is real
Check these before reading a single row. A grid that fails them is not a governance artifact. It is a picture of one.
The first rule is not a consulting preference. UK government project delivery guidance states it flatly: there can only be one accountable person who can be held to account, and that accountability cannot be delegated or shared [R3]. Where a role is formally job-shared, the guidance treats it as a single entity for accountability. The mandatory functional standard goes further and makes defining roles and accountabilities a requirement, including the outcomes each person is responsible for and the person they are accountable to [R4]. A 2026 parliamentary committee reached for the same instrument on a long-running shared services program, recommending a single person be appointed to be in charge, after two independent reviews cited the absence of a single point of ownership [R1].
2What shared ownership costs
The committee count and the auditor's conclusion that what the department considered governance was in fact just a project management structure come from a national audit office performance audit [R5]. The finding that no individual or governance body held the authority to ensure the necessary steps were taken comes from the independent lessons review of the same program [R6]. Separately, 65% of agency directors and fiscal officers on a statewide ERP said it had negatively affected their ability to carry out statutory duties, after the state left the change management plan to the implementer [R9].
National audit work makes the design point directly: good governance means being clear about who has authority and accountability, and that clarity matters most where decisions and accountabilities stretch across a system or a set of organizations [R2]. A client and its integrator are exactly that condition.
3The four letters, defined tightly
Most grids fail because the letters mean different things to different signatories. Print these on the grid itself.
Does the work. Can be several people. The only letter that can repeat in a row without weakening it. Nobody marked responsible means the activity is not resourced.
Answers for the outcome, has authority to decide, carries the consequence. Exactly one. If they cannot direct the people marked responsible, the assignment is wrong and the row fails quietly.
Input required before the work is complete. Asked before, not told after. Keep it disciplined: a row consulted with six parties is a row that will not move.
Told after, one way, no input expected. Where most stakeholders belong. Being generous with I costs nothing. Being generous with C costs weeks.
4The rows, and the ones people forget
Rows are activities, not tasks and not deliverables. An activity has one owner across the whole cutover and appears in the plan as many lines. Write each as a noun phrase. Ten groups cover a cutover, and what matters more than the list is the row inside each group that goes missing.
| Activity group | The row people forget |
|---|---|
| Plan and schedule | Approving a change to the plan after baseline. Usually undefined, so anyone edits |
| Freeze | Granting exceptions during the freeze. This always happens and is almost never assigned |
| Data | Accepting the reconciliation, as distinct from performing it |
| Integrations | Coordinating third parties, sitting between the integration team and the vendor manager |
| Security and access | Removing legacy access. Nobody wants it and auditors always ask |
| Environment | Lifting the change freeze, which decides when normal change resumes |
| Business readiness | Department attestation. Frequently declined by every candidate owner |
| Decisions and gates | Rollback authority. Assumed to sit with the sponsor and rarely confirmed with them |
| Communications | Status during the window to audiences outside the command center |
| Support transition | Tier three definition, and who owns the escalation into it |
On columns: draft from the org chart, then correct it for who is part time, who splits duties, who has a vacancy underneath them. Add the columns nobody asks for, meaning department representatives, the vendor manager, the security function, the service desk that inherits the calls. Those are exactly where the unowned rows end up. Use the sponsor column sparingly, on the go decision, rollback authority and anchor date changes only, because a sponsor accountable for ten rows is accountable for none. And never create a column for a group. Groups decide, individuals answer.
5Keeping the two sides apart
This is the single change that does most of the work, and it is usually agreed quickly once it is framed correctly.
- Functional Lead, holding both organizations
- A row can be marked accountable with two people behind the letter
- Handoffs between organizations are invisible, because they happen inside a cell
- Nobody can tell from the grid which deliverables were contracted
- Disputes settled by memory of who said what in which meeting
- Client Functional Lead and Partner Functional Lead, separately
- Exactly one accountable per row, held by a named individual in one column
- Every handoff appears as an R in one column and a C or A in the other
- Contracted deliverables are visible as partner-column accountable rows
- Disputes settled by reading the row
Frame it as not blurring contractual authority. That wins the argument. Framing it as accountability gets heard as blame, and then you spend a week on feelings instead of on the grid. The harder version of this sits in the audit record: a national audit office found a defence ERP program's governance undermined by conflicts of interest embedded in decision-making arrangements, with contractors involved in decisions relating to their own contracts, and recommended the department avoid contractors participating in decisions on their own contract variations [R8]. If the partner column and the client column are one column, that conflict is not visible.
6A worked cutover grid
| Activity | Sponsor | Client cutover lead | Partner cutover lead | Client track lead | Partner track lead | PMO | Dept rep |
|---|---|---|---|---|---|---|---|
| Author and maintain the cutover plan | I | A | C | R | R | C | |
| Confirm task durations and start conditions | C | C | A | R | I | ||
| Baseline the plan | A | R | C | C | C | R | |
| Approve a change to the plan after baseline | I | A | C | R | C | I | |
| Design the freeze by transaction type | I | A | C | R | C | C | |
| Grant an exception during the freeze | I | A | I | C | I | R | |
| Execute the data load sequence | C | A | I | R | |||
| Reconcile converted data to source | C | R | A | R | C | ||
| Accept the reconciliation result | I | C | I | A | C | I | C |
| Coordinate third-party integration partners | C | C | A | R | I | ||
| Remove legacy system access | C | I | A | C | I | I | |
| Department readiness and attestation | I | C | I | C | R | A | |
| Make the go and no-go decision | A | R | C | C | C | C | C |
| Invoke rollback | A | R | C | I | C | I | I |
The second grid most programs need is support and hypercare, and it is the one most often missing. Support accountability gets assumed rather than assigned, which is why the first week after go-live is spent working out where calls go. The three rows that decide whether the first month works are tier three definition, central versus department resolution, and the managed service boundary.
7The rows that always get contested
Every program contests the same seven. Bring the settling argument into the session rather than discovering the dispute in it.
| Contested row | Why | How to settle it |
|---|---|---|
| Department readiness | Nobody wants thirty departments as their problem, so it lands with change management by default and is not resourced | Split it. Content to change, completion to a named rep per department, aggregate view to the PMO. One row becomes three and all three get owned |
| Data quality remediation | The partner converts what it is given. The client says the partner should have flagged it. Both are partly right | Accountable client side, responsible shared, and a separate partner-accountable row: notifying the client of quality failures within a stated window |
| Third-party coordination | Sits between the integration team, who know the interface, and the vendor manager, who owns the relationship | Accountable to whoever holds the commercial relationship, responsible to whoever knows the interface. Do not merge them |
| Cross-organization decisions | The partner takes work to a point and the client has to decide. Nobody writes down where that point is | Make each decision point its own row with a client-side accountable, and add it to the plan as a dated task |
| Tier three escalation | Everyone assumes it is the partner, the contract usually says otherwise, and nobody reads it until it is needed | Read the agreement in the session. Where it is silent, assign client side and raise the gap commercially |
| Knowledge transfer | Treated as a shared aspiration rather than a deliverable, so it happens informally and incompletely | A partner-side accountable row with acceptance criteria, and a client-side accountable row for accepting that it occurred |
| Post-freeze change tracking | Change requests keep arriving after scope freeze and nobody owns the portfolio view or the regression consequence | One row for the portfolio, PMO accountable. One row for regression scope, test lead accountable |
Two of these have hard public evidence. On data, a published business case risk register for a large council ERP treated data migration volume and quality as mitigated by using a third-party supplier to manage an end-to-end migration cycle [R7], the classic move of transferring the activity while the consequence stays with the client. On readiness, a state legislative evaluation found the state had intended the implementer to develop the change management plan, called that a critical deficiency, and traced generic contract deliverables and one-size-fits-all training straight back to it [R9].
8When both sides claim it: three questions
Not a stalemate to be mediated. Three questions, asked in order. The first clear answer settles it, and you record which one did.
The cost of skipping this shows up in litigation. In a published High Court judgment from a large insurance sector transformation, the release plan expressly assigned the customer its own obligations: product descriptions, sprint decision-making, communication templates, non-standard business rules, all end-user training, and the telephony solution [R10]. When the customer's subject matter experts under-delivered, the customer's own transformation director wrote that the supplier had stepped away and would now position this as a relief event [R10]. That is an unowned client deliverable, seen from a courtroom.
9Grid to plan to runbook
The RACI is the first link in a chain. It is only worth building if the other two get made.
A RACI published once and never revisited is accurate for about six weeks. Set the reopen triggers at publication, because afterwards nobody wants to reopen it.
| Trigger | What to re-check |
|---|---|
| A named person leaves or changes role | Every row where they held A or R. Reassign and re-confirm. Do not inherit by job title |
| Availability materially changes | Rows held by anyone going part time, taking extended absence, or picking up a second engagement |
| New workstream or scope area | Whether it introduces activities the grid does not cover, and which existing rows it changes |
| Phase change | Design, test, cutover and hypercare need different grids. Ownership legitimately moves between them |
| A row fails in practice | The activity did not happen and the accountable party was surprised by it. The most valuable trigger and the least used |
| Rehearsal findings | Every row the rehearsal proved wrong. Rehearsals surface ownership gaps more reliably than any review |
| Contract amendment or change order | Whether the deliverable set moved between the organizations |
10Confirm every accountable row in writing
One step gets skipped more than any other, and it is the one that makes the grid real. Each named accountable party confirms individually, in writing, that they hold the row. Not their manager. Not the track lead on their behalf. Them.
A grid published without that confirmation is a proposal that looks like a decision, and the difference only becomes visible at the moment you needed it to be a decision. The failure mode scales: a 2026 federal audit of shared services adoption found key leadership roles vital to decision-making still unfilled, and two of its four recommendations were simply to fill the named accountable roles and appoint the governance chair [R11]. Unassigned is a decision too.
Public cautionary cases are cited by source and left unnamed in the body so the article stays vendor and organization neutral. Field examples come from real Workday and ERP implementations, fully anonymized. General guidance, not legal advice.