The first few weeks of a VMware migration feel productive. The team connects to vCenter, exports inventory, and begins categorizing workloads. Stakeholder meetings fill the calendar. Momentum is high.
Then something shifts. Around day 60 or 90, progress slows. The easy VMs are mapped, but the hard ones remain. Dependencies surface that were not in the original plan. Application owners stop responding to requests for validation. The pilot that was supposed to complete in week six is still in testing at week ten.
This is the 90-day cliff. Gartner's assessment of large-scale VMware migrations found that projects involving 2,000 or more VMs take 18 to 48 months to complete. It is where VMware migrations go to stall, where timelines slip from months to quarters, and where leadership starts asking whether the project is still viable. Forrester received 175 client guidance session requests about strategic decision-making around VMware between June 2023 and December 2024, reflecting the scale of organizations grappling with this challenge. Understanding why migrations stall at this point is the first step to avoiding that outcome.
Why migrations lose momentum at 90 days
The 90-day mark is not arbitrary. It represents a transition from planning to execution, from easy work to hard work, and from internal effort to external coordination. Several factors converge at this stage.
The easy scope is done
Early migration work focuses on well-documented VMs with clear ownership and simple configurations. These workloads move quickly, creating a sense of progress. But as those are completed, what remains is harder: legacy systems, undocumented dependencies, VMs owned by teams that have moved on.
The work required per VM increases while the VM count decreases. Gartner research VP Julia Palmer counseled against planning to move all workloads off VMware, noting that no rival vendor offers a superior platform and a full migration will take three or more years. Progress metrics that looked strong in month one flatten in month three.
Dependencies compound
During planning, teams identify obvious dependencies. During execution, they discover the dependencies behind those dependencies. A database migration requires application changes, but the application changes require testing environments that do not exist yet.
Each discovered dependency adds work. The project scope expands while the timeline remains fixed. Something has to give.
Stakeholder fatigue sets in
Early in a project, stakeholders are engaged. Meetings are well-attended. Responses to questions come quickly. Over time, attention drifts. Other priorities emerge. The migration becomes one of many demands on stakeholder time.
Response times lengthen. Approvals stall. The migration team waits while stakeholders decide whether to prioritize their requests.
Technical debt surfaces
Production environments accumulate workarounds, temporary configurations, and undocumented changes. These only become visible when migration forces a close examination. Each piece of technical debt requires a decision: remediate before migration, remediate after, or accept the risk.
These decisions take time. Some require investment that was not budgeted. The project slows as the team negotiates scope and resources.
Leadership attention moves elsewhere
Projects that seem on track receive less executive focus. When early phases complete smoothly, leadership assumes the rest will follow. Escalation paths that existed at kickoff are less responsive at day 90.
When the team encounters blockers that require executive intervention, getting attention takes longer. The project waits while leaders context-switch back to a topic they thought was handled.
The costs of a stalled migration
Stalled migrations are expensive in ways that go beyond the direct labor cost of the project team.
VMware renewal pressure. If the migration was planned to complete before a renewal deadline, a stall may force a renewal on unfavorable terms. The cost savings that justified the project evaporate.
Team morale. Engineers working on a stalled project lose motivation. The best people seek assignments on projects that are moving. Knowledge leaves the team.
Credibility damage. A stalled migration becomes evidence that the organization cannot execute complex initiatives. Future projects face more skepticism. Budget approvals become harder.
Technical drift. While the migration is stalled, the VMware environment continues to change. New VMs are deployed. Configurations drift. The analysis done in month one becomes stale by month four.
Opportunity cost. Resources tied up in a stalled migration are unavailable for other initiatives. The organization pays twice: once for the stalled project and again for the delayed alternatives.
Warning signs that a stall is coming
Watch for these indicators before day 90. Early intervention is easier than recovery.
-
Pilot overruns: If the pilot phase takes twice as long as planned, that ratio will likely persist. Investigate why before assuming production waves will be different.
-
Rising remediation backlog: If the list of VMs requiring remediation grows faster than the team resolves it, the timeline is at risk. Address the backlog or adjust expectations.
-
Declining meeting attendance: When key stakeholders stop attending migration meetings or send delegates without decision authority, engagement is fading.
-
Approval delays: Track the time between requesting approval and receiving it. If delays are lengthening, escalation may be necessary.
-
Scope creep requests: Frequent requests to add VMs or change wave assignments indicate that scope was not locked. Unstable scope makes progress measurement impossible.
-
Status report ambiguity: If status reports use vague language like "in progress" or "being reviewed" without specific completion criteria, visibility is insufficient.
How to recover from a stalled migration
If your migration has already stalled, recovery is possible. The approach depends on the root cause.
Rebuild the scope baseline
Stalled migrations often lack a clear, agreed-upon scope. Rebuild consensus on what is in and out. Get explicit sign-off from leadership on the scope and timeline. Make the tradeoffs visible: if the scope stays large, the timeline extends. If the timeline must hold, the scope contracts.
Reduce work in progress
Teams stall when they try to advance too many VMs simultaneously. Limit work in progress to a number the team can actually complete. Finish bundles fully before starting new ones. Completed work builds momentum.
Escalate blockers explicitly
Blockers that have persisted for weeks need escalation. Document each blocker, the owner responsible for resolution, and the date it was identified. Present this list to leadership and ask for intervention.
Do not wait for blockers to resolve themselves. They will not.
Reset stakeholder engagement
Schedule a restart meeting with key stakeholders. Acknowledge that the project has slowed. Present a revised plan with clear asks for each stakeholder. Make commitments mutual: the migration team will deliver on their tasks if stakeholders deliver on theirs.
Inject additional capacity
Sometimes migrations stall because the team is simply undersized for the scope. Consider adding resources, whether internal staff or external partners, to increase throughput. The cost of additional capacity is often less than the cost of extended timeline.
Improve visibility
Leaders cannot help projects they cannot see. Implement dashboards that show real-time progress, blockers, and risk. Make updates available without requiring status meetings. When problems are visible, they are more likely to receive attention.
How VirtualReady prevents the 90-day cliff
ReadyWorks VirtualReady is designed to maintain momentum through the phases where migrations typically stall.
-
Automated data integration keeps the inventory current. Unlike point-in-time exports, the platform synchronizes with vCenter continuously. Technical drift does not invalidate earlier analysis.
-
Dependency mapping surfaces blockers during planning, not execution. Teams know what is hard before they commit to timelines. Scope is realistic from the start.
-
Readiness scoring provides objective progress metrics. Instead of vague status updates, dashboards show exactly which bundles are ready, which are blocked, and why. Progress is measurable at any moment.
-
Workflow automation keeps stakeholders engaged without manual chasing. Approval requests route automatically. Reminders escalate when deadlines approach. The system does the coordination that human project managers struggle to sustain.
-
Executive dashboards give leadership visibility without requiring status meetings. When leaders can see the project status at any time, they engage problems earlier.
The result is a migration that maintains velocity through the hard middle phases, avoiding the cliff that stalls projects run on spreadsheets and manual coordination.
The conversation to have at day 60
If your migration is approaching the 90-day mark, have an honest conversation with your team and leadership now.
Where are we versus plan? Not where do we hope to be, but where are we actually?
What blockers exist that we have not escalated? Surface the issues that have been waiting for someone else to notice.
Is our scope still achievable in our timeline? If not, which gives: scope or timeline?
Do we have the capacity to finish? If not, what additional resources would change the outcome?
Having this conversation at day 60 gives time to adjust. Having it at day 120 means the stall has already cost months.
FAQ
Why do migrations stall around 90 days specifically?
This timeframe reflects the transition from planning to execution, depletion of easy scope, and typical decay in stakeholder engagement. The exact timing varies, but the pattern is consistent.
How do I know if my migration is stalled versus just slow?
Stalled migrations have blockers that are not being actively resolved. Slow migrations have work in progress that is advancing. Check whether blocked items have owners and resolution dates.
Can a stalled migration be restarted without losing work?
Yes. Work completed before the stall remains valid, though analysis may need refresh. A restart focuses on rebuilding momentum and resolving blockers, not redoing completed tasks.
What if leadership is not willing to address blockers?
Document the impact of each blocker on timeline and cost. Present the tradeoff clearly: these blockers will resolve with intervention or the timeline will extend by this amount. Make the choice explicit.
How much time does recovery typically add?
Recovery adds 4 to 12 weeks depending on the depth of the stall. Early intervention minimizes the extension.
One next step
If your VMware migration is approaching the 90-day mark, get an objective assessment of where you stand. Request a VM Accelerator review to establish a clear baseline and identify blockers.
