MSP Client Migration Playbook: Environment Transition Guide
Migration is where onboarding margin goes to die. The MSP onboarding analysis puts migration costs at $5,000-$15,000 per client - the single largest expense in the onboarding cycle. And unlike discovery or stabilization, migration is the phase where scope creep hits hardest because you are deep inside the client’s environment and every problem is visible.
This playbook treats migration as what it actually is: a fixed-scope project with defined boundaries, timeline, and pricing. Not an open-ended commitment bundled into the retainer.
Pre-Migration: Define the Scope (Before You Touch Anything)
The scope comes directly from the paid discovery assessment. If you skipped paid discovery, your migration scope is a guess. Guesses cost money.
The discovery deliverable should give you:
- Complete asset inventory with current state documentation
- Network topology with all subnets, VLANs, and WAN links
- Known issues categorized as: critical (fix during migration), deferred (fix post-stabilization), and out of scope (quote separately)
- A clear list of what gets deployed: RMM, PSA, backup, security stack, MFA, email (if migrating)
From this, build a migration statement of work with explicit inclusions and exclusions.
Migration SOW Template
| Category | In Scope | Out of Scope |
|---|---|---|
| RMM Deployment | Agent install on all inventoried endpoints | New hardware procurement |
| MX record updates, mailbox migration | Archive migration beyond 12 months | |
| Backup | Deploy solution, configure all documented servers | Historical backup data import |
| Security | EDR + DNS filtering + MFA enrollment | Penetration testing, compliance remediation |
| Network | Documentation of current config | Redesign or new hardware |
| Legacy Issues | Items flagged “critical” in assessment | Everything flagged “deferred” or “out of scope” |
That last row is the most important. Legacy issues are the scope creep vector. Without explicit categorization from discovery, every legacy issue becomes an assumed inclusion.
The Migration Sprint: Week by Week
Week 1: Infrastructure Layer
- Deploy RMM agents to all endpoints (workstations, servers, network devices)
- Configure PSA integration and ticket routing
- Deploy backup solution and run initial full backup
- Verify backup integrity with test restore
- Deploy security stack (EDR, DNS filtering, email security gateway)
Checkpoint: All agents reporting, backup verified, security stack active. If any of these are incomplete, do not proceed to Week 2.
Week 2: User Layer
- MFA enrollment for all users (start with admins, then general staff)
- Email migration or MX record cutover (if applicable)
- Printer and peripheral configuration
- VPN or remote access setup and testing
- Client portal provisioning and user access
Checkpoint: All users can log in, email is flowing, printers work, remote access tested. This is the week where ticket volume starts to climb.
Week 3: Documentation and Handoff
- Complete internal documentation in knowledge base
- Run full asset reconciliation (discovery inventory vs deployed inventory)
- Document all passwords, licenses, and vendor contacts
- Conduct client-facing handoff meeting (how to submit tickets, escalation path, SLAs)
- Begin user training sessions
Checkpoint: Documentation complete, client team trained on new tools and processes.
Week 4 (If Needed): Remediation Buffer
- Address critical legacy issues flagged in discovery
- Resolve migration-related tickets
- Fine-tune monitoring thresholds and alert rules
- Clean up temporary configurations
This week is a buffer, not a commitment. Well-scoped migrations with clean environments finish in 2-3 weeks. Messy environments with deferred maintenance need the full 4 weeks.
Cost Control: The Three Rules
Rule 1: Price the migration separately. Typical pricing by environment size:
| Endpoints | Migration Price | Includes |
|---|---|---|
| 10-25 | $5,000-$8,000 | Standard deployment, 2-week timeline |
| 25-75 | $8,000-$15,000 | Full deployment, 3-week timeline |
| 75-150 | $15,000-$25,000 | Enterprise deployment, 4-6 week timeline |
Rule 2: Track hours against estimate. If migration labor exceeds the estimate by 20%, stop and assess. Either the scope expanded (client issue - quote the overage) or the estimate was wrong (your issue - adjust for next time). Do not silently absorb overruns.
Rule 3: “While you’re in there” gets a quote. Every request outside the SOW gets a written quote before work begins. Not a verbal agreement, not a “we’ll figure it out later.” A written number. This is the single most effective margin protection in the migration phase.
The Handoff to Stabilization
Migration is complete when all deployment items in the SOW are verified and documented. It is not complete when the environment is “perfect” - perfect is the enemy of done in this context.
The remaining work - legacy issue remediation, user adaptation, ticket volume normalization - belongs to the stabilization phase. Mixing migration and stabilization into one phase removes the scope boundary and extends the timeline indefinitely.
Check your team’s capacity ceiling before committing to migration timelines. Two concurrent migrations with a 5-person team is manageable. Three is a staffing crisis. Plan the pipeline accordingly, and see how onboarding quality affects retention to understand why rushing migrations to close more deals is a false economy.