Your developer says you should move to the cloud. Your hosting provider keeps sending emails about their legacy platform. Maybe you’re worried about that server sitting under someone’s desk.
Whatever the reason, you’re considering cloud migration. And you’ve heard the horror stories: months of delays, budgets that doubled, systems that broke and stayed broken.
Here’s the thing: most migration disasters aren’t caused by technical problems. They’re caused by skipping steps that seem optional but aren’t.
This checklist is for business owners who aren’t technical but need to make sure their migration doesn’t become a disaster. You won’t learn how to configure AWS—that’s your developer’s job. But you will learn what questions to ask, what red flags to watch for, and which shortcuts to avoid.
TL;DR: The Quick Version
Before you start:
- Know exactly what you’re migrating and why
- Get a realistic cost estimate (not just hosting, but total migration cost)
- Have verified backups of everything
- Define what “success” looks like
During migration:
- Start with something non-critical
- Test thoroughly before switching over
- Have a rollback plan if things go wrong
- Keep the old system running until the new one is proven
After migration:
- Monitor closely for first few weeks
- Don’t delete old systems too fast (see full checklist below)
Common mistakes:
- Underestimating the timeline
- Not accounting for hidden costs
- Skipping the testing phase
- Deleting the old system too soon
What Is Cloud Migration, Really?
Cloud migration is moving your applications, data, and IT processes from wherever they are now (on-premise servers, old hosting providers, that server under the desk) to cloud infrastructure (AWS, Azure, Google Cloud, or similar).
It’s not magic. It’s moving house for your technology. And like moving house, it can be straightforward or chaotic depending on how well you prepare.
What typically gets migrated:
- Websites and web applications
- Databases (customer data, orders, content)
- Email systems
- File storage
- Internal tools and software
- Backups and disaster recovery
What cloud migration services typically include:
- Assessment of your current setup
- Planning the migration approach
- Executing the actual move
- Testing and validation
- Training your team
- Post-migration support
Phase 1: Assessment (Don’t Skip This)
The assessment phase answers one question: What exactly are we dealing with?
This seems obvious, but I’ve seen migrations fail because nobody actually documented what systems existed or how they connected to each other.
The Assessment Checklist
Inventory everything:
| Item | Questions to Answer |
|---|---|
| Applications | What software runs on your servers? Web apps, databases, email, custom tools? |
| Data | How much data? Where is it? How fast does it grow? |
| Integrations | What talks to what? Payment systems, email providers, third-party APIs? |
| Users | Who uses what? Internal team, customers, partners? |
| Dependencies | What breaks if something else breaks? |
Questions to ask your technical team:
- “Can you give me a complete list of every system, application, and database we’re running?”
- “For each one, who uses it and how critical is it to daily operations?”
- “What are the connections between systems? If X goes down, what else breaks?”
- “Are there any known issues with our current setup that migration should fix?”
- “Are there any systems that might be difficult to migrate? Why?”
Red flags during assessment:
- “I’m not sure what that server does” — You need to know before migrating
- “We don’t have documentation” — Create it now, not during migration
- “Only [person who left] understood that system” — Critical knowledge gap
- “We’ve never tested our backups” — Stop and test them immediately
Understanding Your Current Costs
Before you can evaluate whether cloud is cheaper, you need to know what you’re paying now:
- Current hosting fees
- Hardware costs and depreciation
- IT staff time spent on maintenance
- Downtime costs (lost revenue, overtime to fix issues)
- Software licenses tied to current setup
Many businesses discover they don’t actually know their current infrastructure costs. That makes it impossible to evaluate whether cloud is a good deal.
Already running cloud but costs are out of control? See our guide on The Hidden Costs of Cloud for optimization strategies that can cut spending by 20-40%.
Phase 2: Planning
Assessment tells you what you have. Planning tells you how you’ll move it.
Choosing Your Migration Strategy
There are several approaches, each with trade-offs:
Lift and Shift (Rehosting)
- Move applications as-is to cloud servers
- Fastest approach, minimal changes
- May not take advantage of cloud features
- Good for: Quick migrations, applications you’ll replace soon anyway
Replatforming
- Move with some modifications for cloud
- Example: Moving database to managed cloud database service
- Balances speed with some cloud benefits
- Good for: Applications you’ll keep for a while
Refactoring
- Rebuild applications to be “cloud-native”
- Most expensive and time-consuming
- Best performance and scalability
- Good for: Core applications that will run for years
For most small businesses, lift-and-shift or replatforming makes sense. Refactoring is usually overkill unless the application is central to your business and will run for many years.
The Planning Checklist
Timeline planning:
| Phase | Typical Duration | What Happens |
|---|---|---|
| Assessment | 1-2 weeks | Document everything |
| Planning | 1-2 weeks | Design the migration |
| Environment setup | 1-2 weeks | Create cloud infrastructure |
| Migration (per system) | 1-5 days each | Move and test |
| Parallel running | 1-2 weeks | Both systems active |
| Cutover | 1 day | Switch to cloud |
| Post-migration | 2-4 weeks | Monitor and stabilize |
Data volume and app complexity dominate these ranges. A 50GB database migrates in hours; a 5TB database takes days. If your current provider is slow to respond, add time.
Budget planning:
| Cost Category | Typical Range | Notes |
|---|---|---|
| Assessment and planning | €2,000-€8,000 | Often underestimated |
| Migration execution | €3,000-€15,000 | Depends on complexity |
| Cloud setup and configuration | €1,000-€5,000 | Initial infrastructure |
| Data transfer | €100-€2,000 | Depends on volume |
| Testing and validation | €1,000-€5,000 | Don’t skimp here |
| Training | €500-€2,000 | For your team |
| Contingency (15-20%) | Variable | For surprises |
Add ongoing costs:
- Monthly cloud hosting
- Managed services fees
- Monitoring and security tools
Get a personalized estimate: Our Migration Cost Calculator helps you model one-time and ongoing costs based on your specific situation.
Defining Success Criteria
Before you start, agree on what “successful migration” means:
- All applications accessible and functional
- Performance equal to or better than before
- All data migrated and verified
- Security controls in place and tested
- Team trained on new environment
- Monitoring and alerting configured
- Backup and recovery tested
- Costs within X% of estimate
Write these down. Review them during and after migration.
Phase 3: Preparation
This is where you do the work that makes the actual migration go smoothly.
The Preparation Checklist
Backups (Critical):
- Full backup of all data and systems
- Backup tested by actually restoring it
- Backup stored in a location unaffected by migration
- Document the backup and restore procedure
I cannot overstate this: Test your backups before migration. I’ve seen migrations where the backup existed but couldn’t be restored. That’s the same as having no backup.
Is your backup strategy solid? Take our Backup Health Check to find out before you start migrating.
Cloud environment setup:
- Cloud accounts created and secured
- Access controls and permissions configured
- Network architecture designed
- Security groups and firewalls configured
- Monitoring and logging enabled
Documentation:
- Current system documentation complete
- Migration runbook created (step-by-step instructions)
- Rollback procedure documented
- Contact list for all involved parties
- Communication plan for stakeholders
Team preparation:
- Roles and responsibilities assigned
- Migration schedule communicated
- Support contacts identified
- Training scheduled for after migration
Phase 4: Migration Execution
This is the actual moving day. If you’ve done phases 1-3 properly, this should be the smoothest part.
The Execution Checklist
Before starting each migration:
- Confirm backup is complete and verified
- Notify affected users
- Confirm rollback procedure is ready
- Have support contacts available
Migration sequence (recommended order):
- Start with low-risk systems — Marketing website, internal tools
- Then move supporting systems — File storage, development environments
- Then databases — More complex, more risk
- Finally, critical production systems — Customer-facing applications
During migration:
- Follow the runbook step by step
- Document any deviations or issues
- Test each system after migration
- Verify data integrity
- Check performance against benchmarks
After each system migration:
- Functionality testing complete
- Performance acceptable
- Integrations working
- Users can access and use the system
- Monitoring confirming healthy operation
The Parallel Running Period
Don’t immediately turn off your old systems. Run both in parallel:
- Keep old systems running for 1-2 weeks minimum
- Compare behaviour between old and new
- Catch problems before they become permanent
- Have a safe fallback if issues emerge
Only decommission old systems when:
- New systems have been stable for at least a week
- No critical issues discovered
- Team is confident in the new setup
- Data verification complete
Phase 5: Post-Migration
The migration is done, but the work isn’t over. The first few weeks are critical.
The Post-Migration Checklist
First 48 hours:
- Monitor closely for errors and issues
- Check performance metrics
- Verify all integrations working
- Collect user feedback
- Document any issues found
First two weeks:
- Daily check-ins with technical team
- Review costs against estimates
- Identify optimization opportunities
- Complete team training
- Update documentation with changes
First month:
- Performance baseline established
- Cost baseline established
- Outstanding issues resolved
- Old systems safely archived (not deleted)
- Lessons learned documented
After 30 days of stability:
- Old systems can be decommissioned
- Final cost analysis complete
- Migration officially closed
Now that you’re in the cloud: Learn how to optimize your cloud spending before costs get out of control.
A retailer I worked with skipped documenting their integrations during assessment. Two days into migration, their payment system stopped talking to their inventory system. What should have been a 3-day migration became a 2-week scramble. The patterns below will look familiar.
Common Challenges and How to Handle Them
Challenge 1: “It’s Taking Longer Than Expected”
Why it happens: Underestimating complexity, discovering undocumented dependencies, unexpected technical issues.
How to handle it:
- Accept that estimates are estimates, not guarantees
- Build buffer time into your original schedule
- Communicate delays early to stakeholders
- Don’t rush—rushed migrations create more problems
Challenge 2: “It’s Costing More Than Expected”
Why it happens: Hidden costs, scope creep, underestimated complexity, parallel running costs.
How to handle it:
- Budget 20% contingency from the start
- Track costs weekly during migration
- Question new scope additions (“Is this necessary for migration or can it wait?”)
- Accept that some overrun is normal
Challenge 3: “Things Keep Breaking”
Why it happens: Incomplete testing, missed dependencies, configuration differences between environments.
How to handle it:
- Have a clear rollback procedure ready
- Test more thoroughly before cutover
- Keep old systems available longer
- Investigate root causes, don’t just fix symptoms
Challenge 4: “The Team Is Struggling”
Why it happens: New tools, new processes, not enough training, change fatigue.
How to handle it:
- Invest in training before and after migration
- Create documentation and guides
- Be patient—learning curves are normal
- Celebrate milestones to maintain morale
Challenge 5: “Performance Is Worse Than Before”
Why it happens: Incorrect sizing, misconfiguration, network latency, database optimization needed.
How to handle it:
- Compare performance metrics to baseline
- Identify specific bottlenecks
- Don’t assume cloud is automatically faster—it needs tuning
- Consider getting specialist help for optimization
When to Get Expert Help
Handle it yourself (with your developer) if:
- Simple setup (static website, basic database)
- Your developer has cloud experience
- Low complexity, few integrations
- Non-critical systems with low downtime tolerance
Get specialist help if:
- Complex setup with many integrated systems
- Compliance requirements (GDPR, PCI, healthcare)
- Zero-downtime requirement
- Your team lacks cloud experience
- Mission-critical systems with high availability needs
The cost of a specialist is often less than the cost of a failed migration. A botched migration can mean weeks of downtime, lost data, and damaged customer relationships.
Book a consultation if you want an expert assessment before committing to a migration approach.
The Downloadable Checklist
Here’s a summary checklist you can print and use:
Pre-Migration
- Complete system inventory
- Document all integrations and dependencies
- Calculate current infrastructure costs
- Define success criteria
- Choose migration strategy
- Create realistic timeline
- Create realistic budget (with 20% contingency)
Preparation
- Full backup completed and tested
- Cloud environment set up
- Migration runbook created
- Rollback procedure documented
- Team roles assigned
- Stakeholders notified
Execution
- Migrate low-risk systems first
- Test each system after migration
- Verify data integrity
- Document issues and deviations
- Keep old systems running in parallel
Post-Migration
- Monitor closely for first 48 hours
- Daily check-ins for first two weeks
- Collect and address user feedback
- Review costs against estimates
- Complete team training
- Archive old systems after 30 days of stability
Key Takeaways
Cloud migration doesn’t have to be a disaster. Most problems come from skipping steps, not from technical complexity.
The assessment phase matters most. You can’t plan a good migration if you don’t understand what you’re migrating.
Budget for surprises. 20% contingency is standard. Migrations that come in under budget are rare.
Don’t rush. A migration that takes an extra week is better than a migration that breaks your business for a month.
Test your backups. Before you touch anything, confirm you can restore your data if everything goes wrong.
Keep old systems running. Parallel operation costs money but prevents disasters.
Define success before you start. If you don’t know what success looks like, you won’t know when you’ve achieved it.
Next Steps
If you’re considering migration:
- Calculate your migration costs
- Document your current systems (start the assessment)
- Book a consultation if you want expert guidance
If you’ve already migrated:
- Audit your cloud costs to find optimization opportunities
- Review our cloud cost optimization guide
- Set up proper backup and disaster recovery
If you’re still deciding:
- Read our guide to cloud infrastructure architecture
- Understand the IaaS vs PaaS vs SaaS differences to choose the right approach
