Migrate into a new data structure
Data Migration Excellence - Complete Prompting Guide
Overview: Move Fast Without Breaking Things
Data migrations are where careers are made or broken. One wrong move can corrupt years of data, break critical processes, and destroy user trust. This guide teaches you how to use Colby AI to execute flawless migrations that preserve data integrity while transforming your systems.
What You'll Learn:
Why 83% of migrations fail and how to be in the 17%
Advanced mapping strategies for complex transformations
Validation techniques that catch issues before they spread
Rollback strategies that save your reputation
The High Stakes of Data Migration
Migration failures cause:
Data Loss: Irreplaceable information vanished
Process Breakage: Workflows stop working
Trust Erosion: Users lose faith in systems
Revenue Impact: Deals stuck in limbo
Compliance Risk: Audit trail gaps
Understanding Colby's Migration Intelligence
Colby orchestrates migrations with precision:
Data Profiling: Understands source complexity
Intelligent Mapping: Suggests optimal transformations
Validation Rules: Catches issues before migration
Test Execution: Simulates before committing
Error Handling: Manages exceptions gracefully
Rollback Planning: Ensures safe retreat path
The Master Migration Framework
"Plan migration from [source] to [target]:
Scope Definition:
- Objects: [Which entities to migrate]
- Volume: [Record counts]
- Timeframe: [Migration window]
- Dependencies: [Related data]
Data Analysis:
- Source quality assessment
- Required transformations
- Mapping complexity
- Validation requirements
Migration Strategy:
- Approach: [Big bang vs. phased]
- Sequence: [Order of operations]
- Validation: [Check points]
- Rollback: [Recovery plan]
Success Criteria:
- Data completeness
- Accuracy thresholds
- Performance targets
- User acceptance
Risk Assessment:
- Potential failures
- Impact analysis
- Mitigation plans
- Contingencies"
Migration Examples: From Risky to Reliable
❌ Poor Migration Request:
"Move data to new system"
Problems:
No scope
No plan
No validation
Disaster waiting
✅ Basic Migration Request:
"Migrate accounts from old CRM to Salesforce"
Better Because:
Clear source/target
Specific object
Defined direction
✅✅ Good Migration Request:
"Migrate 50,000 accounts from LegacyCRM to Salesforce, map fields, validate data"
Much Better:
Volume specified
Mapping mentioned
Validation included
✅✅✅ Exceptional Migration Request:
"Execute enterprise account migration from LegacyCRM to Salesforce:
Migration Scope:
- Accounts: 50,000 records
- Contacts: 150,000 records
- Opportunities: 25,000 open deals
- Activities: 2 years of history
- Attachments: 500GB documents
Source System Analysis:
- Data quality: 70% complete
- Custom fields: 47 unique
- Relationships: Complex hierarchies
- Duplicates: ~15% estimated
- Legacy codes: Need translation
Field Mapping Matrix:
Complex Transformations:
- Industry codes → Salesforce picklist
- Revenue (string) → Currency field
- Status (1,2,3) → (Active, Inactive, Prospect)
- Country codes → Full names
- Product codes → Product2 lookups
Data Cleansing Requirements:
- Standardize company names
- Validate email formats
- Parse phone numbers
- Geocode addresses
- De-duplicate before migration
Migration Sequence:
1. Reference data (users, products)
2. Account hierarchies (parent first)
3. Accounts (with validation)
4. Contacts (with relationships)
5. Opportunities (preserve stage)
6. Activities (maintain timeline)
7. Attachments (weekend window)
Validation Checkpoints:
- Pre-migration: Source data quality
- Post-mapping: Transformation accuracy
- Post-load: Record counts
- Relationship: Hierarchy integrity
- Business: User acceptance
Rollback Strategy:
- Full backup before start
- Checkpoint after each object
- Rollback scripts prepared
- Communication plan ready
- 48-hour monitoring period
Success Metrics:
- 99.9% data completeness
- Zero data loss
- <5% error rate
- Full audit trail
- User sign-off"
Why This Is Exceptional:
Comprehensive scope definition
Complex transformation handling
Phased approach with checkpoints
Clear success criteria
Risk mitigation built-in
Mastering Different Migration Types
1. System Consolidation Migration
Purpose: Merge multiple systems into one Focus: Conflict resolution and deduplication
Master Prompt:
"Consolidate three regional CRMs into global Salesforce:
Source Systems:
- Americas CRM: 30K accounts
- EMEA System: 25K accounts
- APAC Database: 20K accounts
Overlap Analysis:
- Global accounts in multiple systems
- Different data standards
- Conflicting ownership
- Duplicate contacts
- Currency variations
Master Data Strategy:
- Account matching rules
- Hierarchy establishment
- Owner assignment logic
- Territory mapping
- Data precedence rules
Harmonization Requirements:
- Picklist value mapping
- Currency conversion
- Date format alignment
- Phone standardization
- Address normalization
Conflict Resolution:
- Most recent wins
- Most complete wins
- Regional precedence
- Manual review queue
- Escalation process
Post-Migration:
- Single global view
- Regional access controls
- Unified reporting
- Process alignment
- Training by region"
2. Cloud Migration
Purpose: Move from on-premise to cloud Focus: Architecture transformation
Master Prompt:
"Migrate on-premise CRM to cloud platform:
Current Architecture:
- Database: Oracle 12c
- Custom code: 50K lines
- Integrations: 15 systems
- Storage: 2TB total
- Users: 500 concurrent
Cloud Transformation:
- Database → Cloud native
- Custom code → Configuration
- Point-to-point → API-based
- File storage → Cloud storage
- Desktop → Web-based
Migration Challenges:
Data Challenges:
- Large object handling
- Attachment migration
- Performance testing
- Bandwidth limitations
Code Transformation:
- Business logic extraction
- Workflow recreation
- Validation rule mapping
- Trigger conversion
- API compatibility
Phased Approach:
- Phase 1: Core CRM data
- Phase 2: Integration rebuild
- Phase 3: Custom functionality
- Phase 4: Historical data
- Phase 5: Decommission legacy
Performance Optimization:
- Batch sizing
- Parallel processing
- Off-hours execution
- Incremental sync
- Cache warming"
3. Acquisition Integration
Purpose: Integrate acquired company data Focus: Business continuity and standardization
Master Prompt:
"Integrate acquired company's CRM into our Salesforce:
Acquisition Context:
- Company size: 500 employees
- Customer base: 10K accounts
- Revenue: $50M
- Industry: Complementary
- Timeline: 90 days
Business Requirements:
- Maintain customer relationships
- Preserve opportunity pipeline
- Enable cross-selling
- Unified reporting
- Minimal disruption
Data Mapping Challenges:
Different Business Models:
- Their: Project-based
- Ours: Subscription
- Solution: Hybrid model
Sales Process Variance:
- Their: 6 stages
- Ours: 5 stages
- Mapping: Consolidate similar
Product Catalog:
- Their: 200 SKUs
- Ours: 150 SKUs
- Overlap: 30 products
Integration Strategy:
Week 1-2: Analysis
- Data quality assessment
- Process documentation
- Stakeholder interviews
- System inventory
Week 3-4: Design
- Field mapping
- Process alignment
- User permissions
- Training plan
Week 5-8: Execution
- Test migration
- User training
- Phased cutover
- Hypercare support
Week 9-12: Optimization
- Process refinement
- Report building
- Full integration
- Success metrics"
4. Data Model Transformation
Purpose: Restructure for new requirements Focus: Complex relationship changes
Master Prompt:
"Transform data model for new business requirements:
Current Model Limitations:
- Flat account structure
- Single opportunity type
- Basic contact roles
- Limited product hierarchy
- No partner data
New Model Requirements:
- Account hierarchies (unlimited depth)
- Multiple opportunity types
- Complex buying committees
- Product bundles/solutions
- Partner ecosystem
Transformation Mapping:
Accounts:
- Current: Single level
- New: Parent/child unlimited
- Migration: Build hierarchy from attributes
Opportunities:
- Current: One record type
- New: New/Upsell/Renewal types
- Migration: Classify by criteria
Contacts:
- Current: Basic roles
- New: Influence matrix
- Migration: Enrich role data
Products:
- Current: Flat list
- New: Bundle hierarchy
- Migration: Create relationships
Data Surgery:
- Split combined fields
- Create junction objects
- Establish lookups
- Build roll-ups
- Set defaults
Validation Requirements:
- Hierarchy loops
- Orphaned records
- Relationship integrity
- Business rule compliance
- Performance testing"
Advanced Migration Techniques
Technique 1: Delta Migration Strategy
"Implement incremental migration:
Initial Load:
- Historical data (>1 year)
- Inactive records
- Reference data
- Low-risk migration
Delta Sync:
- Recent changes only
- Active records
- High-priority data
- Continuous sync
Cutover Strategy:
- Freeze source system
- Final delta sync
- Validation checks
- Switch users
- Monitor closely
Benefits:
- Reduced risk
- Faster cutover
- Easy rollback
- Less downtime"
Technique 2: Parallel Run Validation
"Run systems in parallel for validation:
Parallel Period:
- Both systems active
- Dual data entry
- Sync critical data
- Compare outputs
Validation Points:
- Record counts
- Calculation accuracy
- Report matching
- Process outcomes
- User experience
Success Criteria:
- 99% output match
- No critical errors
- User acceptance
- Performance acceptable
- Process compliance
Cutover Decision:
- All criteria met
- Stakeholder approval
- Rollback unnecessary
- Training complete"
Technique 3: Migration Simulation
"Test migration with production-like data:
Simulation Environment:
- Full data volume
- Real complexity
- Actual integrations
- True performance
Test Scenarios:
- Happy path
- Error conditions
- Performance limits
- Rollback procedures
- Recovery processes
Metrics Collection:
- Migration duration
- Error rates
- Resource usage
- Validation accuracy
- User tasks
Optimization:
- Batch size tuning
- Parallel processing
- Error handling
- Performance bottlenecks
- Resource allocation"
Common Migration Mistakes
Mistake 1: Underestimating Complexity
❌ "It's just moving data" ✅ "Plan for transformations, validations, and exceptions"
Mistake 2: No Business Involvement
❌ IT-only project ✅ "Business stakeholders drive requirements and validation"
Mistake 3: Insufficient Testing
❌ Testing with sample data ✅ "Test with full volume and real complexity"
Mistake 4: Poor Communication
❌ Surprise migration ✅ "Communicate early, often, and clearly"
Measuring Migration Success
Technical Metrics:
Data completeness: 99.9%
Error rate: <1%
Performance: Meeting SLAs
Downtime: Minimal
Rollback: Not needed
Business Metrics:
User adoption
Process continuity
Revenue protection
Productivity maintained
Compliance preserved
Pro Tips for Migration Excellence
The Pre-Migration Cleanup
The Measurement Baseline
The Hypercare Period
The Lessons Learned
Explore other Prompting Guides
Search...
⌘K
