Phased vs big-bang go-live
By Emil Björk · Microsoft business apps consultant, Gothenburg
When to phase a Dynamics 365 go-live and when to go big-bang — the trade-offs across complexity, risk, integration burden, and organisational change.
On this page (8)
For any Dynamics 365 implementation beyond the smallest, one of the earliest strategic decisions is whether to roll out all at once (big-bang) or incrementally (phased). Both have credentials and both have failure modes; the right choice depends on the specific shape of the project.
Big-bang
A single cutover: on one date, every legacy system stops accepting new transactions and Dynamics 365 takes over. Common for SMB Business Central implementations and for single-entity F&O rollouts. The advantages: one cutover weekend, no parallel-running maintenance, clean break, single training and change-management push.
The risks: every defect surfaces simultaneously, all to the same support team. If something breaks badly, there's no fallback short of rollback. Pressure on hypercare is enormous. The legacy system is decommissioned (or frozen read-only); recovery options shrink.
Phased by module
Roll out finance first, then operations, then sales, etc. Each phase is its own mini-go-live. Common when the customer is moving from a single legacy ERP and can keep the legacy running for non-migrated modules. The advantages: each go-live is smaller and lower-risk; the team learns between phases; failures contained.
The risks: prolonged dual-running (legacy + Dynamics 365) requires integrations that work both ways during the transition. Period-end close becomes a multi-system exercise. Total elapsed time is longer.
Phased by site or entity
Roll out site A or legal entity 1 first, then expand. Common for multi-site manufacturers or multi-entity groups. The pilot site is the laboratory; lessons feed expansion. The advantages: real-world validation before scaling, lower blast radius, opportunity to refine training and support before expansion.
The risks: until the last site is live, the project isn't finished. Inter-site integrations have to work across both old and new during the rollout. Some scope creep often happens between phases as later sites demand "the same as Site A plus".
Phased by geography
A geographic variant of site phasing, common for global rollouts. Country-specific localizations, languages, holidays, and currencies all become per-phase scope. The most complex variant; only undertaken by mature programs with multi-year horizons.
Choosing
Big-bang for: small-to-medium SMB ERP, time-sensitive contracts, single-site operations, clean replacement of obsolete systems. Phased by module for: large enterprise complexity with mature legacy. Phased by site/geo for: multi-site/global with risk concentration.
Hybrid
Many programs phase the finance go-live (big-bang for the back office) and then phase the operations and CRM rollouts site-by-site. This combines fast core stabilisation with gradual operational coverage.
The discipline
Whichever path, write the rollback plan before you commit. Big-bang rollback is hard but possible; phased rollback is easier but more frequent — both need procedures.
Where to go next
Whichever shape you choose, the mechanics are in cutover planning and the weeks after in hypercare. Phasing decisions are driven by data migration strategy and change management, and the risks of each path belong in the risk register.
Frequently asked questions
When is a big-bang go-live the right choice?
- Small-to-medium ERP implementations, single-site operations, time-sensitive contracts, and clean replacement of an obsolete system — the common shape for Business Central and single-entity F&O rollouts. One cutover weekend, one training push, no dual-running.
What are the ways to phase a rollout?
- By module (finance first, then operations, then sales), by site or legal entity (a pilot site as the laboratory), or by geography (country-specific localisations and languages per phase — the most complex variant).
What is the main cost of phasing?
- Dual-running. Legacy and Dynamics 365 must integrate both ways during the transition, period-end close becomes a multi-system exercise, and total elapsed time is longer. Scope creep between phases is common.
Is there a hybrid approach?
- Yes, and it is common: big-bang the finance go-live to stabilise the back office quickly, then phase operations and CRM rollouts site by site. Whichever path, write the rollback plan before committing.
Related guides
- Build vs buy in the Dynamics 365 ecosystemWhen to customise Dynamics 365 with AL / X++ / Power Platform vs buy an AppSource solution — the trade-offs, the framework, the practical guidance.
- The Success by Design methodologyMicrosoft's current Dynamics 365 implementation methodology — iterative delivery, fit-to-standard, and the role of Solution Blueprint Reviews.
- How to choose the right Dynamics 365 productA practical framework for picking the right Dynamics 365 apps — by company size, industry, complexity, and starting point.
- Architecture decision records for Dynamics 365How ADRs capture the architectural choices in a Dynamics 365 program — what they are, what to record, and the long-term value they provide.
- FastTrack for Dynamics 365Microsoft's FastTrack program for Dynamics 365 — eligibility, what's offered, key engagements, and how to make the most of it.
Browse every guide in Implementation or just Methodology.
Was this helpful?
Signals which guides land and which need work. No account, no comment box — corrections go through the contact page.
Spot something wrong or want a topic covered? Send a correction or a topic request — both are welcome.