Phase 9 of 10 · Launch
Transition
How do we move the work without breaking it?
The big picture
The transition is the highest-risk phase in the life of a GCC. The gap between design and reality is largest, stakeholder confidence is most fragile, and the centre’s long-term credibility is won or lost. It is not a phase you manage; it is a phase you win or lose.
The decisions that matter
- Go/no-go criteria for cutover, as measurable thresholds
- Who can delay cutover, without political penalty
- Hypercare exit criteria, agreed before go-live
- How the sending organisation is supported and incentivised
What good looks like at the end
- A readiness assessment across process, people and organisation
- A completed knowledge transfer ladder, ending in a clean parallel run
- A scripted cutover plan with a rollback trigger
- Agreed hypercare exit criteria and an SME release plan
Who owns it
The GCC Head and programme sponsor jointly; change management as its own funded workstream
When
Wave by wave. Each wave repeats the five stages.
The five-stage transition playbook
| Stage | Critical success factor | Failure mode to avoid |
|---|---|---|
| Readiness assessment | Process stability, supervised execution and governance confirmed | Treating readiness as a document checklist |
| Knowledge transfer | Documentation, shadowing, reverse shadowing, parallel run. No compression. | Cutting reverse shadowing or the parallel run to hit a date |
| Cutover protocol | Scripted window, measurable go/no-go criteria, named rollback trigger | Going live when the criteria are not fully met |
| Stabilisation | Ramp-up curve agreed, issues resolved within five working days, proactive communication | Waiting for performance to improve instead of driving it |
| Hypercare exit | Exit criteria agreed before go-live; exit decided jointly with the business | Letting hypercare drift and dependency set in |
What specialist advisers say
Curated from published work by ISG. Summarised in our own words; follow the link for the full piece.
ISG
Resolve assumptions before signing, write verifiable acceptance criteria for critical deliverables, and validate knowledge transfer with weekly playback sessions scored against an agreed method.
Written for transitions to outsourcing providers; the disciplines carry over to a GCC or BOT transfer.
Making the Move: Top Strategies for Managed Services Transitions, 29 April 2025
In detail
The approach
Test readiness in three dimensions
Process (a process with more than three active variants, documentation written in the last sixty days, or an exception rate above 10% is not ready), people (supervised practice, not reading), and organisation (named escalation contacts, response times, daily reviews).
Climb the knowledge transfer ladder
Documented capture, then at least four weeks of shadowing covering a month-end, then reverse shadowing, then a full parallel run with outputs compared within an agreed tolerance.
Require three supervised cycles
Every person executing a wave one process completes at least three full cycles, trigger to close, under the eye of an experienced SME before cutover.
Script the cutover
One agreed moment of transfer, measurable go/no-go criteria, a 48 to 72 hour scripted window with named owners, a communication plan, and a pre-agreed rollback trigger.
Manage the sending organisation
Explain the why and the timeline, measure SMEs on the quality of knowledge transfer, and give everyone affected clarity on their post-transition role before transfer begins.
Plan hypercare and its end
A named SME on call, a daily thirty-minute review with the business and a weekly programme review, tapering as independence grows. Exit when SLAs hold for two cycles and the business agrees.
Readiness signals
The signs that need attention before a centre moves on from this phase.
- Each wave one process meets the process readiness thresholds
- Everyone on day one has completed three supervised cycles
- A full parallel run matched within tolerance for a complete cycle
- Go/no-go criteria are measurable, and someone is empowered to say no
- Hypercare exit criteria and the SME release plan are agreed
Where organisations go wrong
Documentation written by someone who never watched the work
What works: Shadow live transactions, including month-end and the awkward exceptions.
A go-live date announced before readiness is proven
What works: Make delaying a cutover possible without political penalty.
An unmanaged sending organisation
What works: Give SMEs incentives and role clarity before knowledge transfer starts.
Change management cut to save budget
What works: Fund it as a delivery workstream with its own lead.
Questions for leadership teams
- Has everyone who executes a wave one process on day one completed three supervised cycles under real conditions?
- Who owns the decision to delay cutover, and can they make it without steering committee approval?
- Do the SMEs transferring knowledge have a real incentive to do it well?