Quick answer: There's no single number, because cost depends heavily on scope, systems involved, and organizational complexity. What's consistent across programs is which cost categories get systematically underestimated data migration effort, change management, and post-go-live sustainment chief among them often because they're harder to scope upfront than software licensing.
The Cost Categories That Actually Drive a Budget
Software licensing and infrastructure. The most visible, most heavily scoped cost, and usually the one estimated most accurately since vendors provide clear pricing.
Implementation and configuration. System building, integration, and customization typically scoped well by implementation partners, though scope creep here is common as requirements get clarified mid-program.
Data migration. Consistently underestimated, because the effort required depends on data quality that's often not fully understood until migration work actually begins a legacy system's "clean" data frequently turns out to need far more remediation than assumed at planning time.
Change management. Frequently the first line item cut when budgets tighten, despite being one of the strongest predictors of whether the transformation actually gets adopted. Underinvesting here doesn't reduce the true cost it just shifts it to post-go-live adoption failure.
Governance overhead. Rarely budgeted as its own line item at all, even though the decision rights matrix, RAID management, and phase gate work that prevents governance gaps requires real, dedicated time from senior staff.
Post-go-live sustainment (Harvest). The most commonly missing line item entirely. The Harvest stage extends 12–24 months beyond go-live, and most budgets simply end at go-live, which is a major reason roughly 73% of organizations can't later prove the transformation's ROI.
Why Two Programs of Similar Size Can Have Very Different Budgets
Two transformations with similar scope and headcount can end up with meaningfully different total costs depending almost entirely on data quality going in and how much change management and governance discipline gets built into the plan from day one not on which software vendor gets selected.
Building a Budget That Doesn't Silently Underfund the Back Half
The single most common budgeting mistake is treating go-live as the finish line rather than the midpoint of the Value dimension's work. Building a business case that explicitly budgets for Harvest-stage benefits tracking, not just implementation, is the fastest way to avoid the ROI-measurement gap most organizations fall into.
Read how to build a defensible business case →
Frequently Asked Questions
Why is data migration cost so hard to estimate upfront?
Because the true condition of legacy data duplicates, inconsistent formats, missing fields is often not fully visible until migration and validation work actually begins, well after initial budgeting is complete.
Should change management have its own budget line, or be folded into implementation?
It should have its own explicit line item. When it's folded into implementation, it tends to get quietly deprioritized whenever the technical work runs over budget, which is a common pattern in underperforming transformations.
How much of the total budget should go toward post-go-live sustainment?
There's no universal percentage, but the more relevant point is that it should be budgeted as a distinct, planned line item covering 12–24 months post-go-live not left unbudgeted, which is the far more common and costly mistake.
Instagram